US6981040B1 - Automatic, personalized online information and product services - Google Patents

Automatic, personalized online information and product services Download PDF

Info

Publication number
US6981040B1
US6981040B1 US09597975 US59797500A US6981040B1 US 6981040 B1 US6981040 B1 US 6981040B1 US 09597975 US09597975 US 09597975 US 59797500 A US59797500 A US 59797500A US 6981040 B1 US6981040 B1 US 6981040B1
Authority
US
Grant status
Grant
Patent type
Prior art keywords
user
product
document
model
documents
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active, expires
Application number
US09597975
Inventor
Yochai Konig
Roy Twersky
Michael R. Berthold
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Personalized User Model PUM
Original Assignee
UTOPY Inc
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Grant date
Family has litigation

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06NCOMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N99/00Subject matter not provided for in other groups of this subclass
    • G06N99/005Learning machines, i.e. computer in which a programme is changed according to experience gained by the machine itself during a complete run
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/30Information retrieval; Database structures therefor ; File system structures therefor
    • G06F17/30861Retrieval from the Internet, e.g. browsers
    • G06F17/30864Retrieval from the Internet, e.g. browsers by querying, e.g. search engines or meta-search engines, crawling techniques, push systems
    • G06F17/30867Retrieval from the Internet, e.g. browsers by querying, e.g. search engines or meta-search engines, crawling techniques, push systems with filtering and personalisation
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network-specific arrangements or communication protocols supporting networked applications
    • H04L67/20Network-specific arrangements or communication protocols supporting networked applications involving third party service providers
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network-specific arrangements or communication protocols supporting networked applications
    • H04L67/26Push based network services
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network-specific arrangements or communication protocols supporting networked applications
    • H04L67/30Network-specific arrangements or communication protocols supporting networked applications involving profiles
    • H04L67/306User profiles
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network-specific arrangements or communication protocols supporting networked applications
    • H04L67/36Network-specific arrangements or communication protocols supporting networked applications involving the display of network or application conditions affecting the network application to the application user
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04NPICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00Selective content distribution, e.g. interactive television, VOD [Video On Demand]
    • H04N21/40Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/43Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network, synchronizing decoder's clock; Client middleware
    • H04N21/442Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
    • H04N21/44213Monitoring of end-user related data
    • H04N21/44222Monitoring of user selections, e.g. selection of programs, purchase activity

Abstract

A method for providing automatic, personalized information services to a computer user includes the following steps: transparently monitoring user interactions with data during normal use of the computer; updating user-specific data files including a set of user-related documents; estimating parameters of a learning machine that define a User Model specific to the user, using the user-specific data files; analyzing a document to identify its properties; estimating the probability that the user is interested in the document by applying the document properties to the parameters of the User Model; and providing personalized services based on the estimated probability. Personalized services include personalized searches that return only documents of interest to the user, personalized crawling for maintaining an index of documents of interest to the user; personalized navigation that recommends interesting documents that are hyperlinked to documents currently being viewed; and personalized news, in which a third party server customized its interaction with the user. The User Model includes continually-updated measures of user interest in words or phrases, web sites, topics, products, and product features. The measures are updated based on both positive examples, such as documents the user bookmarks, and negative examples, such as search results that the user does not follow. Users are clustered into groups of similar users by calculating the distance between User Models.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of U.S. Provisional Application No. 60/173,392 filed Dec. 28, 1999, which is herein incorporated by reference.

FIELD OF THE INVENTION

This invention relates generally to methods for personalizing a user's interaction with information in a computer network. More particularly, it relates to methods for predicting user interest in documents and products using a learning machine that is continually updated based on actions of the user and similar users.

BACKGROUND ART

The amount of static and dynamic information available today on the Internet is staggering, and continues to grow exponentially. Users searching for information, news, or products and services are quickly overwhelmed by the volume of information, much of it useless and uninformative. A variety of techniques have been developed to organize, filter, and search for information of interest to a particular user. Broadly, these methods can be divided into information filtering techniques and collaborative filtering techniques.

Information filtering techniques focus on the analysis of item content and the development of a personal user interest profile. In the simplest case, a user is characterized by a set of documents, actions regarding previous documents, and user-defined parameters, and new documents are characterized and compared with the user profile. For example, U.S. Pat. No. 5,933,827, issued to Cole et al., discloses a system for identifying new web pages of interest to a user. The user is characterized simply by a set of categories, and new documents are categorized and compared with the user's profile. U.S. Pat. No. 5,999,975, issued to Kittaka et al., describes an online information providing scheme that characterizes users and documents by a set of attributes, which are compared and updated based on user selection of particular documents. U.S. Pat. No. 6,006,218, issued to Breese et al., discloses a method for retrieving information based on a user's knowledge, in which the probability that a user already knows of a document is calculated based on user-selected parameters or popularity of the document. U.S. Pat. No. 5,754,939, issued to Herz et al., discloses a method for identifying objects of interest to a user based on stored user profiles and target object profiles. Other techniques rate documents using the TFIDF (term frequency, inverse document frequency) measure. The user is represented as a vector of the most informative words in a set of user-associated documents. New documents are parsed to obtain a list of the most informative words, and this list is compared to the user's vector to determine the user's interest in the new document.

Existing information filtering techniques suffer from a number of drawbacks. Information retrieval is typically a two step process, collection followed by filtering; information filtering techniques personalize only the second part of the process. They assume that each user has a personal filter, and that every network document is presented to this filter. This assumption is simply impractical given the current size and growth of the Internet; the number of web documents is expected to reach several billion in the next few years. Furthermore, the dynamic nature of the documents, e.g., news sites that are continually updated, makes collection of documents to be filtered later a challenging task for any system. User representations are also relatively limited, for example, including only a list of informative words or products or user-chosen parameters, and use only a single mode of interaction to make decisions about different types of documents and interaction modes. In addition, information filtering techniques typically allow for extremely primitive updating of a user profile, if any at all, based on user feedback to recommended documents. As a user's interests change rapidly, most systems are incapable of providing sufficient personalization of a user's experience.

Collaborative filtering methods, in contrast, build databases of user opinions of available items, and then predict a user opinion based on the judgments of similar users. Predictions typically require offline data mining of very large databases to recover association rules and patterns; a significant amount of academic and industrial research is focussed on developing more efficient and accurate data mining techniques. The earliest collaborative filtering systems required explicit ratings by the users, but existing systems are implemented without the user's knowledge by observing user actions. Ratings are inferred from, for example, the amount of time a user spends reading a document or whether a user purchases a particular product. For example, an automatic personalization method is disclosed in B. Mobasher et al., “Automatic Personalization Through Web Usage Mining,” Technical Report TR99-010, Department of Computer Science, Depaul University, 1999. Log files of documents requested by users are analyzed to determine usage patterns, and online recommendations of pages to view are supplied to users based on the derived patterns and other pages viewed during the current session.

Recently, a significant number of web sites have begun implementing collaborative filtering techniques, primarily for increasing the number and size of customer purchases. For example, Amazon.com™ has a “Customers Who Bought” feature, which recommends books frequently purchased by customers who also purchased a selected book, or authors whose work is frequently purchased by customers who purchased works of a selected author. This feature uses a simple “shopping basket analysis”; items are considered to be related only if they appear together in a virtual shopping basket. Net Perceptions, an offshoot of the GroupLens project at the University of Minnesota, is a company that provides collaborative filtering to a growing number of web sites based on data mining of server logs and customer transactions, according to predefined customer and product clusters.

Numerous patents disclose improved collaborative filtering systems. A method for item recommendation based on automated collaborative filtering is disclosed in U.S. Pat. No. 6,041,311, issued to Chislenko et al. Similarity factors are maintained for users and for items, allowing predictions based on opinions of other users. In an extension of standard collaborative filtering, item similarity factors allow predictions to be made for a particular item that has not yet been rated, but that is similar to an item that has been rated. A method for determining the best advertisements to show to users is disclosed in U.S. Pat. No. 5,918,014, issued to Robinson. A user is shown a particular advertisement based on the response of a community of similar users to the particular advertisement. New ads are displayed randomly, and the community interest is recorded if enough users click on the ads. A collaborative filtering system using a belief network is disclosed in U.S. Pat. No. 5,704,317, issued to Heckerman et al., and allows automatic clustering and use of non-numeric attribute values of items. A multi-level mindpool system for collaborative filtering is disclosed in U.S. Pat. No. 6,029,161, issued to Lang et al. Hierarchies of users are generated containing clusters of users with similar properties.

Collaborative filtering methods also suffer from a number of drawbacks, chief of which is their inability to rate content of an item or incorporate user context. They are based only on user opinions; thus an item that has never been rated cannot be recommended or evaluated. Similarly, obscure items, which are rated by only a few users, are unlikely to be recommended. Furthermore, they require storage of a profile for every item, which is unfeasible when the items are web pages. New items cannot be automatically added into the database. Changing patterns and association rules are not incorporated in real time, since the data mining is performed offline. In addition, user clusters are also static and cannot easily be updated dynamically.

Combinations of information filtering and collaborative filtering techniques have the potential to supply the advantages provided by both methods. For example, U.S. Pat. No. 5,867,799, issued to Lang et al., discloses an information filtering method that incorporates both content-based filtering and collaborative filtering. However, as with content-based methods, the method requires every document to be filtered as it arrives from the network, and also requires storage of a profile of each document. Both of these requirements are unfeasible for realistically large numbers of documents. An extension of this method, described in U.S. Pat. No. 5,983,214, also to Lang et al., observes the actions of users on content profiles representing information entities. Incorporating collaborative information requires that other users have evaluated the exact content profile for which a rating is needed.

In summary, none of the existing prior art methods maintain an adaptive content-based model of a user that changes based on user behavior, allow for real-time updating of the model, operate during the collection stage of information retrieval, can make recommendations for items or documents that have never been evaluated, or model a user based on different modes of interaction.

OBJECTS AND ADVANTAGES

Accordingly, it is a primary object of the present invention to provide a method of personalizing user interaction with network documents that maintains an adaptive content-based profile of the user.

It is another object of the invention to incorporate into the profile user behavior during different modes of interaction with information, thus allowing for cross-fertilization. Learning about the user interests in one mode benefits all other modes.

It is a further object of the invention to provide a method that jointly models the user's information needs and product needs to provide stronger performance in both modes.

It is an additional object of the invention to provide a method that personalizes both the collection and filtering stages of information retrieval to manage efficiently the enormous number of existing web documents.

It is another object of the invention to provide a method for predicting user interest in an item that incorporates the opinions of similar users without requiring storage and maintenance of an item profile.

It is a further object of the invention to provide an information personalization method that models the user as a function independent of any specific representation or data structure, and represents the user interest in a document or product independently of any specific user information need. This approach enables the addition of new knowledge sources into the user model.

It is an additional object of the present invention to provide a method based on Bayesian statistics that updates the user profile based on both negative and positive examples.

It is a further object of the invention to model products by analyzing all relevant knowledge sources, such as press releases, reviews, and articles, so that a product can be recommended even if it has never been purchased or evaluated previously.

SUMMARY

These objects and advantages are attained by a computer-implemented method for providing automatic, personalized information services to a user. User interactions with a computer are transparently monitored while the user is engaged in normal use of the computer, and monitored interactions are used to update user-specific data files that include a set of documents associated with the user. Parameters of a learning machine, which define a User Model specific to the user, are estimated from the user-specific data files. Documents that are of interest and documents that are not of interest to the user are treated distinctly in estimating the parameters. The parameters are used to estimate a probability P(u|d) that a document is of interest to the user, and the estimated probability is then used to provide personalized information services to the user.

The probability is estimated by analyzing properties of the document and applying them to the learning machine. Documents of multiple distinct media types of analyzed, and identified properties include: the probability that the document is of interest to users who are interested in particular topics, a topic classifier probability distribution, a product model probability distribution, product feature values extracted from the document, the document author, the document age, a list of documents linked to the document, the document language, number of users who have accessed the document, number of users who have saved the document in a favorite document list, and a list of users previously interested in the document. All properties are independent of the particular user. The product model probability distribution, which indicates the probability that the document refers to particular products, is obtained by applying the document properties to a product model, a learning machine with product parameters characterizing particular products. These product parameters are themselves updated based on the document properties and on the product model probability distribution. Product parameters are initialized from a set of documents associated with each product.

User interactions are monitored during multiple distinct modes of user interaction with network data, including a network searching mode, network navigation mode, network browsing mode, email reading mode, email writing mode, document writing mode, viewing “pushed” information mode, finding expert advice mode, and product purchasing mode. Based on the monitored interactions, parameters of the learning machine are updated. Learning machine parameters define various user-dependent functions of the User Model, including a user topic probability distribution representing interests of the user in various topics, a user product probability distribution representing interests of the user in various products, a user product feature probability distribution representing interests of the user in various features of each of the various products, a web site probability distribution representing interests of the user in various web sites, a cluster probability distribution representing similarity of the user to users in various clusters, and a phrase model probability distribution representing interests of the user in various phrases. Some of the user-dependent functions can be represented as information theory based measures representing mutual information between the user and either phrases, topics, products, features, or web sites. The product and feature distributions can also be used to recommend products to the user.

The User Model is initialized from documents provided by the user, a web browser history file, a web browser bookmarks file, ratings by the user of a set of documents, or previous product purchases made by the user. Alternatively, the User Model may be initialized by selecting a set of predetermined parameters of a prototype user selected by the user. Parameters of the prototype user are updated based on actions of users similar to the prototype user. The User Model can be modified based on User Model modification requests provided by the user. In addition, the user can temporarily use a User Model that is built from a set of predetermined parameters of a profile selected by the user.

Distances between users are calculated to determine similar users, who are clustered into clusters of similar users. Parameters defining the User Model may include the calculated distances between the User Model and User Models of users within the user's cluster. Users may also be clustered based on calculated relative entropy values between User Models of multiple users.

A number of other probabilities can be calculated, such as a posterior probability P(u|d,q) that the document is of interest to the user, given a search query submitted by the user. Estimating the posterior probability includes estimating a probability that the query is expressed by the user with an information need contained in the document. In addition, the probability P(u|d,con) that the document is of interest to the user during a current interaction session can be calculated. To do so, P(u,con|d)/P(con|d) is calculated, where con represents a sequence of interactions during the current interaction session or media content currently marked by the user. A posterior probability P(u|d,q,con) that the document is of interest to the user, given a search query submitted during a current interaction session, can also be calculated.

A variety of personalized information services are provided using the estimated probabilities. In one application, network documents are crawled and parsed for links, and probable interest of the user in the links is calculated using the learning machine. Links likely to be of interest to the user are followed. In another application, the user identifies a document, and a score derived from the estimated probability is provided to the user. In an additional application, the user is provided with a three-dimensional map indicating user interest in each document of a hyperlinked document collection. In a further application, an expert user is selected from a group of users. The expert user has an expert User Model that indicates a strong interest in a document associated with a particular area of expertise. Another application includes parsing a viewed document for hyperlinks and separately estimating for each hyperlink a probability that the linked document is of interest to the user. In a further application, user interest information derived from the User Model is sent to a third party web server that then customizes its interaction with the user. Finally, a set of users interested in a document is identified, and a range of interests for the identified users is calculated.

BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a schematic diagram of a computer system in which the present invention is implemented.

FIG. 2 is a block diagram of a method of the present invention for providing personalized product and information services to a user.

FIG. 3 is a schematic diagram of knowledge sources used as inputs to the User Model and resulting outputs.

FIGS. 4A–4E illustrate tables that store different components and parameters of the User Model.

FIG. 5A illustrates a cluster tree containing clusters of users similar to a particular user.

FIG. 5B is a table that stores parameters of a user cluster tree.

FIG. 6A illustrates a preferred cluster tree for implementing fuzzy or probabilistic clustering.

FIG. 6B is a table that stores parameters of a user fuzzy cluster tree.

FIG. 7 illustrates a portion of a topic tree.

FIG. 8 is a table that stores nodes of the topic tree of FIG. 7.

FIG. 9 is a table that stores the names of clusters having the most interest in nodes of the topic tree of FIG. 7, used to implement the topic experts model.

FIG. 10 illustrates a portion of a product tree.

FIG. 11 is a table that stores nodes of the product tree of FIG. 10.

FIG. 12A is a table that stores feature values of products of the product tree of FIG. 10.

FIG. 12B is a table that stores potential values of product features associated with intermediate nodes of the product tree of FIG. 10.

FIG. 13 is a schematic diagram of the method of initializing the User Model.

FIG. 14 illustrates the user recently accessed buffer, which records all user interactions with documents.

FIG. 15A is a table for storing sites that are candidates to include in the user site distribution.

FIG. 15B is a table for storing words that are candidates to include in the user word distribution.

FIG. 16 is a table that records all products the user has purchased.

FIG. 17 is a schematic diagram of the method of applying the User Model to new documents to estimate the probability of user interest in the document.

FIG. 18 is a block diagram of the personal crawler application of the present invention.

FIG. 19 is a block diagram of the personal search application of the present invention.

FIG. 20 is a block diagram of the personal navigation application of the present invention.

FIG. 21 is a block diagram of the document barometer application of the present invention.

FIG. 22 is a schematic diagram of the three-dimensional map application of the present invention.

DETAILED DESCRIPTION

Although the following detailed description contains many specifics for the purposes of illustration, anyone of ordinary skill in the art will appreciate that many variations and alterations to the following details are within the scope of the invention. Accordingly, the following preferred embodiment of the invention is set forth without any loss of generality to, and without imposing limitations upon, the claimed invention.

The present invention, referred to as Personal Web, provides automatic, personalized information and product services to a computer network user. In particular, Personal Web is a user-controlled, web-centric service that creates for each user a personalized perspective and the ability to find and connect with information on the Internet, in computer networks, and from human experts that best matches his or her interests and needs. A computer system 10 implementing Personal Web 12 is illustrated schematically in FIG. 1. Personal Web 12 is stored on a central computer or server 14 on a computer network, in this case the Internet 16, and interacts with client machines 18, 20, 22, 24, 26 via client-side software. Personal Web 12 may also be stored on more than one central computers or servers that interact over the network. The client-side software may be part of a web browser, such as Netscape Navigator or Microsoft Internet Explorer, configured to interact with Personal Web 12, or it may be distinct from but interacting with a client browser. Five client machines are illustrated for simplicity, but Personal Web 12 is intended to provide personalized web services for a large number of clients simultaneously.

For all of the typical interactions that a user has with a computer network, such as the world wide web, Personal Web 12 provides a personalized version. Personal Web 12 stores for each user a User Model 13 that is continuously and transparently updated based on the user's interaction with the network, and which allows for personalization of all interaction modes. The User Model represents the user's information and product interests; all information that is presented to the user has been evaluated by the User Model to be of interest to the user. The User Model allows for cross fertilization; that is, information that is learned in one mode of interaction is used to improve performance in all modes of interaction. The User Model is described in detail below.

Five examples of personalized interaction modes provided by the present invention are illustrated in FIG. 1. However, it is to be understood that the present invention provides for personalization of all modes, and that the following examples in no way limit the scope of the present invention. Personal Web is active during all stages of information processing, including collection, retrieval, filtering, routing, and query answering.

Client 18 performs a search using Personal Web 12 by submitting a query and receiving personalized search results. The personal search feature collects, indexes, and filters documents, and responds to the user query, all based on the user profile stored in the User Model 13. For example, the same query (e.g., “football game this weekend” or “opera”) submitted by a teenager in London and an adult venture capitalist in Menlo Park returns different results based on the personality, interests, and demographics of each user. By personalizing the collection phase, the present invention does not require that all network documents be filtered for a particular user, as does the prior art.

Client 20 browses the web aided by Personal Web 12. In browsing mode, the contents of a web site are customized according to the User Model 13. Personal Web interacts with a cooperating web site by supplying User Model information, and a web page authored in a dynamic language (e.g., DHTML) is personalized to the user's profile. In navigation mode, a personal navigation aid suggests to the user relevant links within the visited site or outside it given the context, for example, the current web page and previously visited pages, and knowledge of the user profile.

Client 22 illustrates the find-an-expert mode of Personal Web 12. The user supplies an expert information or product need in the form of a sample web page or text string, and Personal Web 12 locates an expert in the user's company, circle of friends, or outside groups that has the relevant information and expertise, based on the expert's User Model The located expert not only has the correct information, but presents it in a manner of most interest to the user, for example, focussing on technical rather than business details of a product.

Client 24 uses the personal pushed information mode of Personal Web 12. Personal Web 12 collects and presents personal information to a user based on the User Model 13. The pushed information is not limited to a fixed or category or topic, but includes any information of interest to the user. In communities, organizations, or group of users, the pushed information can include automatic routing and delivery of newly created documents that are relevant to the users.

Finally, client 26 illustrates the product recommendation mode of Personal Web 12. The user submits a query for information about a product type, and Personal Web 12 locates the products and related information that are most relevant to the user, based on the User Model 13. As described further below, product information is gathered from all available knowledge sources, such as product reviews and press releases, and Personal Web 12 can recommend a product that has never been purchased or rated by any users.

All of the above features of Personal Web 12 are based on a User Model 13 that represents user interests in a document or product independently of any specific user information need, i.e., not related to a specific query. The User Model 13 is a function that is developed and updated using a variety of knowledge sources and that is independent of a specific representation or data structure. The underlying mathematical framework of the modeling and training algorithms discussed below is based on Bayesian statistics, and in particular on the optimization criterion of maximizing posterior probabilities. In this approach, the User Model is updated based on both positive and negative training examples. For example, a search result at the top of the list that is not visited by the user is a negative training example.

The User Model 13, with its associated representations, is an implementation of a learning machine. As defined in the art, a learning machine contains tunable parameters that are altered based on past experience. Personal Web 12 stores parameters that define a User Model 13 for each user, and the parameters are continually updated based on monitored user interactions while the user is engaged in normal use of a computer. While a specific embodiment of the learning machine is discussed below, it is to be understood that any model that is a learning machine is within the scope of the present invention.

The present invention can be considered to operate in three different modes: initialization, updating or dynamic learning, and application. In the initialization mode, a User Model 13 is developed or trained based in part on a set of user-specific documents. The remaining two modes are illustrated in the block diagram of FIG. 2. While the user is engaged in normal use of a computer, Personal Web 12 operates in the dynamic learning mode to transparently monitor user interactions with data (step 30) and update the User Model 13 to reflect the user's current interests and needs. This updating is performed by updating a set of user-specific data files in step 32, and then using the data files to update the parameters of the User Model 13 in step 34. The user-specific data files include a set of documents and products associated with the user, and monitored user interactions with data. Finally, Personal Web 12 applies the User Model 13 to unseen documents, which are first analyzed in step 36, to determine the user's interest in the document (step 38), and performs a variety of services based on the predicted user interest (step 40). In response to the services provided, the user performs a series of actions, and these actions are in turn monitored to further update the User Model 13.

The following notation is used in describing the present invention. The user and his or her associated representation are denoted with u, a user query with q, a document with d, a product or service with p, a web site with s, topic with t, and a term, meaning a word or phrase, with w. The term “document” includes not just text, but any type of media, including, but not limited to, hypertext, database, spreadsheet, image, sound, and video. A single document may have one or multiple distinct media types. Accordingly, the set of all possible documents is D, the set of all users and groups is U, the set of all products and services is P, etc. The user information or product need is a subset of D or P. Probability is denoted with P, and a cluster of users or of clusters with c, with which function semantics are used. For example, c(c(u)) is the cluster of clusters in which the user u is a member (“the grandfather cluster”). Note that an explicit notation of world knowledge, such as dictionaries, atlases, and other general knowledge sources, which can be used to estimate the various posterior probabilities, is omitted.

A document classifier is a function whose domain is any document, as defined above, and whose range is the continuous interval [0, 1]. For example, a document classifier may be a probability that a document d is of interest to a particular user or a group of users. Specific document classifiers of the present invention are obtained using the User Model 13 and Group Model. The User Model 13 represents the user interest in a document independent of any specific user information need. This estimation is unique to each user. In strict mathematical terms, given a user u and a document d, the User Model 13 estimates the probability P(u|d). P(u|d) is the probability of the event that the user u is interested in the document d, given everything that is known about the document d. This classifier is extended to include P(u|d,con), the probability that a user is interested in a given document based on a user's current context, for example, the web pages visited during a current interaction session.

The Group or Cluster Model is a function that represents the interest level of a group of users in a document independently of any specific information need. For example, for the group of users c(u), the mathematical notation of this probability, which is determined by applying the Group Model to a document d, is P(c(u)|d).

A schematic diagram of the User Model is shown in FIG. 3, which illustrates the various knowledge sources (in circles) used as input to the User Model. The knowledge sources are used to initialize and update the User Model, so that it can accurately take documents and generate values of user interest in the documents, given the context of the user interaction. Note that some of the knowledge sources are at the individual user level, while others refer to aggregated data from a group of users, while still others are independent of all users. Also illustrated in FIG. 3 is the ability of the User Model to estimate a user interest in a given product, represented mathematically as the interest of a user in a particular document, given that the document describes the product: P(user|document, product described=p). As explained further below, the long-term user interest in a product is one of many probabilities incorporated into the computation of user interest in all documents, but it can also be incorporated into estimation of a current user interest in a product.

Beginning at the bottom left of FIG. 3, User Data and Actions include all user-dependent inputs to the User Model, including user browser documents, user-supplied documents, other user-supplied data, and user actions, such as browsing, searching, shopping, finding experts, and reading news. Data and actions of similar users are also incorporated into the User Model by clustering all users into a tree of clusters. Clustering users allows estimation of user interests based on the interests of users similar to the user. For example, if the user suddenly searches for information in an area that is new to him or her, the User Model borrows characteristics of User Models of users with similar interests. Topic classifiers are used to classify documents automatically into topics according to a predefined topic tree. Similarly, product models determine the product or product categories, if any, referred to by a document. Product models also extract relevant feature of products from product-related documents. The topic experts input provides input of users with a high interest in a particular topic, as measured by their individual User Models. Finally, the User Model incorporates world knowledge sources that are independent of all users, such as databases of company names, yellow pages, thesauri, dictionaries, and atlases.

User Model Representations

Given the inputs shown in FIG. 3, the User Model is a function that may be implemented with any desired data structure and that is not tied to any specific data structure or representation. The following currently preferred embodiment of abstract data structures that represent the User Model 13 is intended to illustrate, but not limit, the User Model of the present invention. Some of the structures hold data and knowledge at the level of individual users, while others store aggregated data for a group or cluster of users. Initialization of the various data structures of the User Model is described in the following section; the description below is of the structures themselves.

User-dependent inputs are represented by components of the User Model shown in FIGS. 4A–4E. These inputs are shown as tables for illustration purposes, but may be any suitable data structure. The user-dependent components include an informative word or phrase list, a web site distribution, a user topic distribution, a user product distribution, and a user product feature distribution. Each of these user-dependent data structures can be thought of as a vector of most informative or most frequent instances, along with a measure representing its importance to the user.

The informative word and phrase list of FIG. 4A contains the most informative words and phrases found in user documents, along with a measure of each informative phrase or word's importance to the user. As used herein, an “informative phrase” includes groups of words that are not contiguous, but that appear together within a window of a predefined number of words. For example, if a user is interested in the 1999 Melissa computer virus, then the informative phrase might include the words “virus,” “Melissa,” “security,” and “IT,” all appearing within a window of 50 words. The sentence “The computer virus Melissa changed the security policy of many IT departments” corresponds to this phrase.

In addition to the words and phrases, the list contains the last access time of a document containing each word or phrase and the total number of accessed documents containing the words. One embodiment of the informative measure is a word probability distribution P(w|u) representing the interest of a user u in a word or phrase w, as measured by the word's frequency in user documents. Preferably, however, the informative measure is not simply a measure of the word frequency in user documents; common words found in many documents, such as “Internet,” provide little information about the particular user's interest. Rather, the informative measure should be high for words that do not appear frequently across the entire set of documents, but whose appearance indicates a strong likelihood of the user's interest in a document. A preferred embodiment uses the TFIDF measure, described in Ricardo Baeza-Yates and Berthier Ribeiro-Neto, Modern Information Retrieval, Addison Wesley, 1999, in which TF stands for term frequency, and IDF stands for inverse document frequency. Mathematically, if fu,w denotes the frequency of the word w in user u documents, and Dw denotes the number of documents containing the word w, then the importance of a word w to a user u is proportional to the product fu,w·D/Dw.

A more preferred embodiment of the measure of each word's importance uses a mathematically sound and novel implementation based on information theory principles. In particular, the measure used is the mutual information between two random variables representing the user and the word or phrase. Mutual information is a measure of the amount of information one random variable contains about another; a high degree of mutual information between two random variables implies that knowledge of one random variable reduces the uncertainty in the other random variable.

For the present invention, the concept of mutual information is adapted to apply to probability distributions on words and documents. Assume that there is a document in which the user's interest must be ascertained. The following two questions can be asked: Does the phrase p appear in the document?; and Is the document of interest to the user u? Intuitively, knowing the answer to one of the questions reduces the uncertainty in answering the other question. That is, if the word w appears in a different frequency in the documents associated with the user u from its frequency in other documents, it helps reduce the uncertainty in determining the interest of user u in the document.

Through the concept of mutual information, information theory provides the mathematical tools to quantify this intuition in a sound way. For a detailed explanation, see T. Cover and J. Thomas, Elements of Information Theory, Wiley, 1991. In this embodiment of the informative measure, two indicator variables are defined. Iw has a value of 1 when the word w appears in a web document and 0 when it does not, and Iu has a value of 1 when a web document is of interest to the user u and 0 when it does not. The mutual information between the two random variables Iw and Iu is defined as: I ( I w ; I u ) = i w I w i u I u P ( i w , i u ) log 2 P ( i w , i u ) P ( i w ) P ( i u )

The probabilities in this formula are computed over a set of documents of interest to the user and a set of documents not of interest to the user. For example, consider a set of 100 documents of interest to the user, and a set of 900 documents not of interest to the user. Then P(iu=1)=0.1, and P(iu=0)=0.9. Assume that in the combined set of 1000 documents, 150 contain the word “Bob.” Then P(iw=1)=0.15, and P(iw=0)=0.85. In addition, assume that “Bob” appears in all 100 of the documents of interest to the user. P(iw,iu) has the following four values:

iu iw P(iw, iu)
0 0 850/1000
0 1  50/1000
1 0  0/1000
1 1 100/1000

Using the above formula, the mutual information between the user and word Bob is: I ( I Bob ; I user ) = 850 / 1000 log [ 850 / 1000 / ( 0.85 * 0.9 ) ] + 50 / 1000 log [ 50 / 1000 / ( 0.15 * 0.9 ) ] + 0 / 1000 log [ 0 / 1000 / ( 0.1 * 0.85 ) ] + 100 / 1000 log [ 100 / 100 / ( 0.15 * 0.1 ) ] = 0.16 .

Mutual information is a preferred measure for selecting the word and phrase list for each user. The chosen words and phrases have the highest mutual information.

The remaining User Model representations are analogously defined using probability distributions or mutual information. The web site distribution of FIG. 4B contains a list of web sites favored by the user along with a measure of the importance of each site. Given the dynamic nature of the Internet, in which individual documents are constantly being added and deleted, a site is defined through the first backslash (after the www). For example, the uniform resource locator (URL) http://www.herring.com/companies/2000 . . . is considered as www.herring.com. Sites are truncated unless a specific area within a site is considered a separate site; for example, www.cnn.com/health is considered to be a different site than www.cnn.com/us. Such special cases are decided experimentally based on the amount of data available on each site and the principles of data-driven approaches, described in Vladimir S. Cherkassky and Filip M. Mulier, Learning from Data: Concepts, Theory, and Methods, in Adaptive and Learning Systems for Signal Processing, Communications and Control, Simon Haykin, series editor, Wiley & Sons, March, 1998. Each site has an importance measure, either a discrete probability distribution, P(s|u), representing the interest of user u in a web site s, or the mutual information metric defined above, I(Is; Iu), representing the mutual information between the user u and a site s. The web site distribution also contains the last access time and number of accesses for each site.

FIG. 4C illustrates the user topic distribution, which represents the interests of the user in various topics. The user topic distribution is determined from a hierarchical, user-independent topic model, for example a topic tree such as the Yahoo directory or the Open Directory Project, available at http://dmoz.org/. Each entry in the tree has the following form:

Computers\Internet\WWW\Searching the Web\Directories\Open Directory Project\

where the topic following a backslash is a child node of the topic preceding the backslash. The topic model is discussed in more detail below.

For each node of the topic tree, a probability is defined that specifies the user interest in the topic. Each level of the topic model is treated distinctly. For example, for the top level of the topic model, there is a distribution in which
P(t u |u)+P(t l |u)=1,
where tl represents the top level of topics and is the same set of topics for each user, e.g., technology, business, health, etc. P (tl|u) is the sum of the user probabilities on all top level topics. For each topic level, tu represents specific interests of each user that are not part of any common interest topics, for instance family and friends' home pages. For lower topic levels, every node in the tree is represented in the user topic distribution by a conditional probability distribution. For example, if the Technology node splits into Internet, Communication, and Semiconductors, then the probability distribution is of the form:
P(Internet|u,Technology)+P(Communication|u,Technology)+P(Semiconductors|u,Technology)+P(tu |u,Technology)=1

Rather than probabilities, the mutual information metric defined above may be used; I(It; Iu) represents the mutual information between the user u and the topic t. An exemplary data structure shown in FIG. 4C for storing the user topic distribution contains, for each topic, the topic parent node, informative measure, last access time of documents classified into the topic, and number of accesses of documents classified into the topic. Note that the User Model contains an entry for every topic in the tree, some of which have a user probability or mutual information of zero.

The user product distribution of FIG. 4D represents the interests of the user in various products, organized in a hierarchical, user-independent structure such as a tree, in which individual products are located at the leaf nodes of the tree. The product taxonomy is described in further detail below. The product taxonomy is similar to the topic tree. Each entry in the tree has the following form:

    • Consumer Electronics\Cameras\Webcams\3Com HomeConnect\
      where a product or product category following a backslash is a child node of a product category preceding the backslash.

For each node of the product model, a probability is defined that specifies the user interest in that particular product or product category. Each level of the product model is treated distinctly. For example, for the top level of the product hierarchy, there is a distribution in which
P(p l |u)=1,
where pl represents the top level of product categories and is the same for each user, e.g., consumer electronics, computers, software, etc. For lower product category levels, every node in the tree is represented in the user product distribution by a conditional probability distribution. For example, if the Cameras node splits into Webcams and Digital Cameras, then the probability distribution is of the form:
P(Webcams|u,Cameras)+P(Digital Cameras|u,Cameras)=1

Rather than probabilities, the mutual information metric defined above may be used. Then I(Ip; Iu) represents the mutual information between the user u and the product or product category p. An exemplary data structure for storing the user product distribution contains, for each product, the product ID, product parent node, user probability, last purchase time of the product, number of product purchases, last access time of documents related to the product, and number of related documents accessed.

For each product or category on which the user has a nonzero probability, the User Model contains a user product feature distribution on the relevant features, as shown in FIG. 4E. Each product category has associated with it a list of features, and the particular values relevant to the user are stored along with a measure of the value's importance, such as a probability P(f|u,p) or mutual information measure I(If; Iu). For example, Webcams have a feature Interface with possible values Ethernet (10BaseT), Parallel, PC Card, serial, USB, and TV. Probability values of each feature sum to one; that is,
P(Ethernet|u,Interface,Webcam)+P(Parallel|u,Interface,Webcam)+P(PCCard|u,Interface,Webcam)+P(serial|u,Interface,Webcam)+P(USB|u,Interface, Webcam)+P(TV|u,Interface,Webcam)=1.

User probability distributions or mutual information measures are stored for each feature value of each node. Note that there is no user feature value distribution at the leaf nodes, since specific products have particular values of each feature.

Finally, user-dependent components of the User Model include clusters of users similar to the user. Users are clustered into groups, forming a cluster tree. One embodiment of a user cluster tree, shown in FIG. 5A, hard classifies users into clusters that are further clustered. Each user is a member of one and only one cluster. For example, Bob is clustered into a cluster c(u), which is further clustered into clusters of clusters, until the top level cluster is reached c(U). The identity of the user's parent cluster and grandfather cluster is stored as shown in FIG. 5B, and information about the parent cluster is used as input into the User Model. As described below, clusters are computed directly from User Models, and thus need not have a predefined semantic underpinning.

Preferably, the User Model does not user hard clustering, but rather uses soft or fuzzy clustering, also known as probabilistic clustering, in which the user belongs to more than one cluster according to a user cluster distribution P(c(u)). FIG. 6A illustrates fuzzy clusters in a cluster hierarchy. In this case, Bob belongs to four different clusters according to the probability distribution shown. Thus Bob is most like the members of cluster C4, but still quite similar to members of clusters C1, C2, C3, and C4. Fuzzy clustering is useful for capturing different interests of a user. For example, a user may be a small business owner, a parent of a small child, and also an avid mountain biker, and therefore need information for all three roles. Probabilistic clustering is described in detail in the Ph.D. thesis of Steven J. Nowlan, “Soft Competitive Adaptation: Neural Network Learning Algorithms Based on Fitting Statistical Mixtures,” School of Computer Science, Carnegie Mellon University, Pittsburgh, Pa., 1991. A suitable data structure for representing fuzzy clusters is shown in FIG. 6B. Each row stores the cluster or user ID, one parent ID, and the cluster probability, a measure of similarity between the cluster or user and the parent cluster.

Note that all elements of an individual User Model for a user u also apply to a cluster of users c(u). Thus for each cluster, a Group Model is stored containing an informative word list, a site distribution, a topic distribution, a group product distribution, and a group product feature distribution, each with appropriate measures. For example, P(p|c(u)) represents the interest of a cluster c(u) in various products p.

The user-dependent User Model representations also include a user general information table, which records global information describing the user, such as the User ID, the number of global accesses, the number of accesses within a recent time period, and pointers to all user data structures.

Other knowledge sources of the User Model are independent of the user and all other users. Topic classifiers are used to classify documents into topics according to a predefined topic tree, an example of which is illustrated in FIG. 7. A variety of topic trees are available on the web, such as the Yahoo directory or Open Directory Project (www.dmoz.org). A topic classifier is a model similar to the user model that estimates the probability that a document belongs to a topic. Every node on the topic tree has a stored topic classifier. Thus the set of all topic classifiers computes a probability distribution of all of the documents in the set of documents D among the topic nodes. For example, the topic classifier in the root node in FIG. 7 estimates the posterior probabilities P(t|d), where t represents the topic of document d and is assigned values from the set {Arts, Business, Health, News, Science, Society}. Similarly, the topic classifier for the Business node estimates the posterior probability P(t|d, Business), where t represents the specific topic of the document d within the Business category. Mathematically, this posterior probability is denoted P(t(d)=Business\Investing\|t(d)=Business, d), which represents the probability that the subtopic of the document d within Business is Investing, given that the topic is Business. The topic tree is stored as shown in FIG. 8, a table containing, for each node, the topic ID, depth level, topic parent ID, number of child nodes, and topic ID of the child nodes.

The topic experts model estimates the probability that a document is of interest to users who are interested in a particular topic, independent of any specific user information need. Each node of the topic tree has, in addition to a topic classifier, a corresponding topic expert function. Note that the topic classifier and topic expert function are independent; two documents can be about investing, but one of high interest to expert users and the other of no interest to expert users. The topic expert model can be considered an evaluation of the quality of information in a given document. The assumption behind the topic experts model is that the degree of interest of a user in a given topic is his or her weight for predicting the quality or general interest level in a document classified within the particular topic. Obviously there are outliers to this assumption, for example, novice users. However, in general and averaged across many users, this measure is a good indicator of a general interest level in a document. For every topic in the tree, a list of the N clusters with the most interest in the topic based on the cluster topic distribution is stored. The cluster topic distribution is similar to the user topic distribution described above, but is averaged over all users in the cluster. An exemplary data structure for storing the topic experts model is shown in FIG. 9.

Finally, a product model is stored for every node of a product taxonomy tree, illustrated in FIG. 10. Examples of product taxonomy trees can be found at www.cnet.com and www.productopia.com, among other locations. In any product taxonomy tree, the leaf nodes, i.e., the bottom nodes of the tree, correspond to particular products, while higher nodes represent product categories. Product models are similar to topic classifiers and User Models, and are used to determine whether a document is relevant to a particular product or product category. Thus a product model contains a list of informative words, topics, and sites. The set of all product models computes a probability distribution of all of the documents in the set of documents D among the product nodes. For example, the product model in the root node in FIG. 10 estimates the posterior probabilities P(p|d), where p represents the product referred to in document d and is assigned values from the set {Consumer Electronics, Computers, Software}. Similarly, the product model for the Consumer Electronics node estimates the posterior probability P(p|d, Consumer Electronics), where p represents the product category of the document d within the Consumer Electronics category. Mathematically, this posterior probability is denoted P(p(d)=Consumer Electronics\CD Players\|p(d)=Consumer Electronics, d), which represents the probability that the subproduct category of the document d within Consumer Electronics is CD Players, given that the product category is Consumer Electronics. The product tree is stored as shown in FIG. 11, a table containing, for each node, the topic ID, depth level, topic parent ID, number of child nodes, and topic ID of the child nodes.

Each node of the product tree has an associated product feature list, which contains particular descriptive features relevant to the product or category. Nodes may have associated feature values; leaf nodes, which represent specific products, have values of all relevant product features. Product feature lists are determined by a human with knowledge of the domain. However, feature values may be determined automatically form relevant knowledge sources as explained below.

For example, in the product tree of FIG. 10, CD Players is the parent node of the particular CD players Sony CDP-CX350 and Harman Kardon CDR2. The product category CD Players has the following features: Brand, CD Capacity, Digital Output, Plays Minidisc, and Price Range. Each feature has a finite number of potential feature values; for example, CD Capacity has potential feature values 1 Disc, 1–10 Discs, 10–50 Discs, or 50 Discs or Greater. Individual products, the child nodes of CD Players, have one value of each feature. For example, the Sony CDP-CX350 has a 300 disc capacity, and thus a feature value of 50 Discs or Greater.

Some product features are relevant to multiple product categories. In this case, product features propagate as high up the product tree as possible. For example, digital cameras have the following product features: PC Compatibility, Macintosh Compatibility, Interfaces, Viewfinder Type, and Price Range. Webcams have the following product features: PC Compatibility, Macintosh Compatibility, Interfaces, Maximum Frames per Second, and Price Range. Common features are stored at the highest possible node of the tree; thus features PC Compatibility, Macintosh Compatibility, and Interfaces are stored at the Cameras node. The Digital Cameras node stores only product feature Viewfinder Type, and the Webcams node stores only product feature Maximum Frames per Second.

Note that product feature Price Range is common to CD Players and Cameras, and also Personal Minidiscs, and thus is propagated up the tree and stored at node Consumer Electronics.

Individual products at leaf nodes inherit relevant features from all of their ancestor nodes. For example, Kodak CD280 inherits the feature Viewfinder Type from its parent; PC Compatibility, Macintosh Compatibility, and Interfaces from its grandparent; and Price Range from its great-grandparent. A product feature list is stored as shown in FIG. 12A, and contains, for each product ID, the associated feature and its value. All potential feature values are stored in a product feature value list, as shown in FIG. 12B.

The system also includes a document database that indexes all documents D. The document database records, for each document, a document ID, the full location (the URL of the document), a pointer to data extracted from the document, and the last access time of the document by any user. A word database contains statistics of each word or phrase from all user documents. The word database contains the word ID, full word, and word frequency in all documents D, used in calculating informative measures for individual users and clusters.

Initialization of User Model

The User Model is initialized offline using characterizations of user behavior and/or a set of documents associated with the user. Each data structure described above is created during initialization. In other words, the relevant parameters of the learning machine are determined during initialization, and then continually updated online during the update mode.

In one embodiment, the user documents for initializing the User Model are identified by the user's web browser. Most browsers contain files that store user information and are used to minimize network access. In Internet Explorer, these files are known as favorites, cache, and history files. Most commercial browsers, such as Netscape Navigator, have equivalent functionality; for example, bookmarks are equivalent to favorites. Users denote frequently-accessed documents as bookmarks, allowing them to be retrieved simply by selection from the list of bookmarks. The bookmarks file includes for each listing its creation time, last modification time, last visit time, and other information. Bookmarks of documents that have changed since the last user access are preferably deleted from the set of user documents. The Internet Temporary folder contains all of the web pages that the user has opened recently (e.g., within the last 30 days). When a user views a web page, it is copied to this folder and recorded in the cache file, which contains the following fields: location (URL), first access time, and last access time (most recent retrieval from cache). Finally, the history file contains links to all pages that the user has opened within a set time period.

Alternatively, the user supplies a set of documents, not included in any browser files, that represent his or her interests. The User Model can also be initialized from information provided directly by the user. Users may fill out forms, answer questions, or play games that ascertain user interests and preferences. The user may also rate his or her interest in a set of documents provided.

User documents are analyzed as shown in FIG. 13 to determine initial parameters for the various functions of the User Model. A similar analysis is used during updating of the User Model. Note that during updating, both documents that are of interest to the user and documents that are not of interest to the user are analyzed and incorporated into the User Model. The process is as follows. In a first step 82, the format of documents 80 is identified. In step 84, documents 80 are parsed and separated into text, images and other non-text media 88, and formatting. Further processing is applied to the text, such as stemming and tokenization to obtain a set of words and phrases 86, and information extraction. Through information extraction, links 90 to other documents, email addresses, monetary sums, people's names, and company names are obtained. Processing is performed using natural language processing tools such as LinguistX® and keyword extraction tools such as Thing Finder™, both produced by Inxight (www.inxight.com). Further information on processing techniques can be found in Christopher D. Manning and Hinrich Schutze, Foundations of Statistical Natural Language Processing, MIT Press, 1999. Additional processing is applied to images and other non-text media 88. For example, pattern recognition software determines the content of images, and audio or speech recognition software determines the content of audio. Finally, document locations 94 are obtained.

Parsed portions of the documents and extracted information are processed to initialize or update the user representations in the User Model. In step 96, user informative words or phrases 98 are obtained from document words and phrases 86. In one embodiment, a frequency distribution is obtained to calculate a TFIDF measure quantifying user interest in words 98. Alternatively, mutual information is calculated between the two indicator variables Iw and Iu as explained above. The set of informative words 98 contains words with the highest probabilities or mutual information.

In step 100, the topic classifiers are applied to all extracted information and portions of documents 80 to obtain a probability distribution P(t|d) for each document on each node of the topic tree. As a result, each node has a set of probabilities, one for each document, which is averaged to obtain an overall topic node probability. The average probabilities become the initial user topic distribution 102. If desired, mutual information between the two indicator variables Is and Iu can be determined as explained above.

Similarly, in step 104, product models are applied to all extracted information from documents 80 to classify documents according to the product taxonomy tree. From user purchase history 105, additional product probabilities are obtained. Probabilities for each node are combined, weighting purchases and product-related documents appropriately, to obtain a user product distribution 106. Note that only some of documents 80 contain product-relevant information and are used to determine the user product distribution 106. Product models return probabilities of zero for documents that are not product related.

The user product feature distribution 108 can be obtained from different sources. If a user has a nonzero probability for a particular product node, then the feature distribution on that node is obtained from its leaf nodes. For example, if one of the user documents was classified into Kodak DC280 and another into Nikon Coolpix 950, then the user product feature distribution for the Digital Cameras node has a probability of 0.5 for the feature values corresponding to each camera. Feature value distributions propagate throughout the user product feature distributions. For example, if the two cameras are in the same price range, $300–$400, then the probability of the value $300–$400 of the feature Price Range is 1.0, which propagates up to the Consumer Electronics node (assuming that the user has no other product-related documents falling within Consumer Electronics).

Alternatively, product feature value distributions are obtained only from products that the user has purchased, and not from product-related documents in the set of user documents. Relevant feature values are distributed as high up the tree as appropriate. If the user has not purchased a product characterized by a particular feature, then that feature has a zero probability. Alternatively, the user may explicitly specify his or her preferred feature values for each product category in the user product distribution. User-supplied information may also be combined with feature value distributions obtained from documents or purchases.

Document locations 94 are analyzed (step 110) to obtain the user site distribution 112. Analysis takes into account the relative frequency of access of the sites within a recent time period, weighted by factors including how recently a site was accessed, whether it was kept in the favorites or bookmarks file, and the number of different pages from a single site that were accessed. Values of weighting factors are optimized experimentally using jackknifing and cross-validation techniques described in H. Bourlard and N. Morgan, Connectionist Speech Recognition: A Hybrid Approach, Kluwer Academic Publishers, 1994.

Note that there is typically overlap among the different representations of the User Model. For example, a news document announcing the release of a new generation of Microsoft servers has relevant words Microsoft and server. In addition, it is categorized within the product taxonomy under Microsoft servers and the topic taxonomy under computer hardware. This document may affect the user's word list, product distribution, and topic distribution.

After the User Models are initialized for all users, cluster membership can be obtained. Clusters contain users with a high degree of similarity of interests and information needs. A large number of clustering algorithms are available; for examples, see K. Fukunaga, Statistical Pattern Recognition, Academic Press, 1990. As discussed above, users are preferably soft clustered into more than one cluster. Preferably, the present invention uses an algorithm based on the relative entropy measure from information theory, a measure of the distance between two probability distributions on the same event space, described in T. Cover and J. Thomas, Elements of Information Theory, Chapter 2, Wiley, 1991. Clustering is unsupervised. That is, clusters have no inherent semantic significance; while a cluster might contain users with a high interest in mountain biking, the cluster tree has no knowledge of this fact.

In a preferred embodiment, the relative entropy between two User Model distributions on a fixed set of documents Dsample is calculated. Dsample is chosen as a good representation of the set of all documents D. Distributions of similar users have low relative entropy, and all pairs of users within a cluster have relative entropy below a threshold value. The User Model of each user is applied to the documents to obtain a probability of interest of each user in each document in the set. The relative entropy between two user distributions for a single document is calculated for each document in the set, and then summed across all documents.

The exact mathematical computation of the relative entropy between two users is as follows. An indicator variable Iu,d is assigned to 1 when a document d is of interest to a user u and 0 when it is not. For two users u1 and u2 and for any document d, the relative entropy between the corresponding distributions is: D ( I u1 , d I u2 , d ) = i I P ( i u1 , d ) log 2 P ( i u1 , d ) P ( i u2 , d )

For example, if P(u1|d)=0.6 and P(u2|d)=0.9, then
D(I u1,d ∥I u2,d)=0.4 log (0.4/0.1)+0.6 log (0.6/0.9).

The relative entropy can be converted to a metric D′ that obeys the triangle inequality:
D′(I 1 ∥I 2)=0.5*(D(I1 ∥I 2)+D(I 2 ∥I 1)).

For any two users u 1 and u2, and for each document in Dsample, the metric D′ is computed between the corresponding indicator variable distributions on the document. The values for all document are summed, and this sum is the distance metric for clustering users. This distance is defined as: Distance ( u 1 , u 2 ) = d j D sample D 2 ( I u1 , d j I u2 , d j ) .

An alternative clustering algorithm computes the relative entropy between individual user distributions in the User Model, for example, between all informative word lists, site distributions, etc., of each user. The equations are similar to those above, but compute relative entropy based on indicator variables such as Iu,w, which is assigned a value of 1 when a word w is of interest to a user u. The calculated distances between individual user distributions on words, sites, topics, and products are summed to get an overall user distance. This second algorithm is significantly less computationally costly than the preferred algorithm above; selection of an algorithm depends on available computing resources. In either case, relative entropy can also be computed between a user and cluster of users.

Each cluster has a Group or Cluster Model that is analogous to a User Model. Cluster Models are generated by averaging each component of its members' User Models. When fuzzy clusters are used, components are weighted by a user's probability of membership in the cluster.

In some cases, initialization is performed without any user-specific information. A user may not have a large bookmarks file or cache, or may not want to disclose any personal information. For such users, prototype users are supplied. A user can choose one or a combination of several prototype User Models, such as the technologist, the art lover, and the sports fan. Predetermined parameters of the selected prototype user are used to initialize the User Model. Users can also opt to add only some parameters of a prototype user to his or her existing User Model by choosing the prototype user's distribution of topics, words, sites, etc. Note that prototype users, unlike clusters, are semantically meaningful. That is, prototype users are trained on a set of documents selected to represent a particular interest. For this reason, prototype users are known as “hats,” as the user is trying on the hat of a prototype user.

Users can also choose profiles on a temporary basis, for a particular session only. For example, in a search for a birthday present for his or her teenage daughter, a venture capitalist from Menlo Park may be interested in information most probably offered to teenagers, and hence may choose a teenage girl profile for the search session.

User-independent components are also initialized. The topic classifiers are trained using the set of all possible documents D. For example, D may be the documents classified by the Open Directory Project into its topic tree. Topic classifiers are similar to a User Model, but with a unimodal topic distribution function (i.e., a topic model has a topic distribution value of 1 for itself and 0 for all other topic nodes). The set of documents associated with each leaf node of the topic tree is parsed and analyzed as with the user model to obtain an informative word list and site distribution. When a topic classifier is applied to a new document, the document's words and location are compared with the informative components of the topic classifier to obtain P(t|d). This process is further explained below with reference to computation of P(u|d). Preferably, intermediate nodes of the tree do not have associated word list and site distributions. Rather, the measures for the word list and site distribution of child nodes are used as input to the topic classifier of their parent nodes. For example, the topic classifier for the Business node of the topic tree of FIG. 7 has as its input the score of the site of the document to be classified according to the site distributions of the topic models of its child nodes, Employment, Industries, and Investing. The classifier can be any non-linear classifier such as one obtained by training a Multilayer Perceptron (MLP) using jackknifing and cross-validation techniques, as described in H. Bourlard and N. Morgan, Connectionist Speech Recognition: A Hybrid Approach, Kluwer Academic Publishers, 1994. It can be shown that a MLP can be trained to estimate posterior probabilities; for details, see J. Hertz, A. Krogh, R. Palmer, Introduction to The Theory of Neural Computation, Addison-Wesley, 1991.

The topic experts model is initialized by locating for every node in the topic tree the N clusters that are of the same depth in the user cluster tree as the user, and that have the highest interest in the topic, based on their cluster topic distribution. The cluster topic distribution P(t|c(u)) is simply an average of the user topic distribution P(t|u) for each user in the cluster. The topic experts model is used to determine the joint probability that a document and the topic under consideration are of interest to any user, P(t,d). Using Bayes' rule, this term can be approximated by considering the users of the N most relevant clusters. P ( t , d ) = i N P ( c i t , d ) P ( t d ) P ( d )

The topic experts model is, therefore, not a distinct model, but rather an ad hoc combination of user and cluster topic distributions and topic models.

Product models are initialized similarly to User Models and topic classifiers. Each leaf node in the product tree of FIG. 10 has an associated set of documents that have been manually classified according to the product taxonomy. These documents are used to train the product model as shown for the User Model in FIG. 13. As a result, each leaf node of the product tree contains a set of informative words, a topic distribution, and a site distribution. Each node also contains a list of features relevant to that product, which is determined manually. From the documents, values of the relevant features are extracted automatically using information extraction techniques to initialize the feature value list for the product. For example, the value of the CD Capacity is extracted from the document. Information extraction is performed on unstructured text, such as HTML documents, semi-structured text, such as XML documents, and structured text, such as database tables. As with the topic model, a nonlinear function such as a Multilayer Perceptron is used to train the product model.

Preferably, as for topic classifiers, intermediate nodes of the product tree do not have associated word lists, site distributions, and topic distributions. Rather, the measures for the word list, site distribution, and topic distribution of child nodes are used as input to the product models of their parent nodes. Alternatively, each parent node may be trained using the union of all documents of its child nodes.

Updating the User Model

The User Model is a dynamic entity that is refined and updated based on all user actions. User interactions with network data are transparently monitored while the user is engaged in normal use of his or her computer. Multiple distinct modes of interaction of the user are monitored, including network searching, network navigation, network browsing, email reading, email writing, document writing, viewing pushed information, finding expert advice, product information searching, and product purchasing. As a result of the interactions, the set of user documents and the parameters of each user representation in the User Model are modified.

While any nonlinear function may be used in the User Model (e.g., a Multilayer Perceptron), a key feature of the model is that the parameters are updated based on actual user reactions to documents. The difference between the predicted user interest in a document or product and the actual user interest becomes the optimization criterion for training the model.

Through his or her actions, the user creates positive and negative patterns. Positive examples are documents of interest to a user: search results that are visited following a search query, documents saved in the user favorites or bookmarks file, web sites that the user visits independently of search queries, etc. Negative examples are documents that are not of interest to the user, and include search results that are ignored although appear at the top of the search result, deleted bookmarks, and ignored pushed news or email. Conceptually, positive and negative examples can be viewed as additions to and subtractions from the user data and resources.

Information about each document that the user views is stored in a recently accessed buffer for subsequent analysis. The recently accessed buffer includes information about the document itself and information about the user's interaction with the document. One possible implementation of a buffer is illustrated in FIG. 14; however, any suitable data structure may be used. The recently-accessed buffer contains, for each viewed document, a document identifier (e.g., its URL); the access time of the user interaction with the document; the interaction type, such as search or navigation; the context, such as the search query; and the degree of interest, for example, whether it was positive or negative, saved in the bookmarks file, how long the user spent viewing the document, or whether the user followed any links in the document. Additional information is recorded for different modes of interaction with a document as discussed below.

A metric is determined for each document to indicate whether it is a positive, negative or neutral event; this metric can potentially be any grade between 0 and 1, where 0 is a completely negative event, 1 is a completely positive event, and 0.5 is a neutral event. Previous user interactions may be considered in computing the metric; for example, a web site that the user accesses at a frequency greater than a predetermined threshold frequency is a positive example. After each addition to or subtraction from the set of user documents, the document is parsed and analyzed as for the User Model initialization. Extracted information is incorporated into the User Model.

Because the User Model is constantly and dynamically updated, applying the initialization process for each update is inefficient. Preferably, incremental learning techniques are used to update the User Model. Efficient incremental learning and updating techniques provide for incorporating new items into existing statistics, as long as sufficient statistics are recorded. Details about incremental learning can be found in P. Lee, Bayesian Statistics, Oxford University Press, 1989.

After a document stored in the recently accessed buffer is parsed, parsed portions are stored in candidate tables. For example, FIGS. 15A and 15B illustrate a user site candidate table and user word candidate table. The user site candidate table holds sites that are candidates to move into the user site distribution of FIG. 4B. The site candidate table stores the site name, i.e., the URL until the first backslash, except for special cases; the number of site accesses; and the time of last access. The user word candidate table holds the words or phrases that are candidates to move into the user informative word list of FIG. 4A. It contains a word or phrase ID, alternate spellings (or misspellings) of the word, an informative grade, and a time of last access.

Negative examples provide words, sites, and topics that can be used in several ways. The measure of any item obtained from the negative example may be reduced in the user distribution. For example, if the negative example is from a particular site that is in the user site distribution, then the probability or mutual information of that site is decreased. Alternatively, a list of informative negative items may be stored. The negative items are obtained from negative examples and are used to reduce the score of a document containing negative items.

Documents are added to the buffer during all user modes of interaction with the computer. Interaction modes include network searching, network navigation, network browsing, email reading, email writing, document writing, viewing “pushed” information, finding expert advice, and product purchasing. Different types of information are stored in the buffer for different modes. In network searching, search queries are recorded and all search results added to the buffer, along with whether or not a link was followed and access time for viewed search results. In network browsing, the user browses among linked documents, and each document is added to the buffer, along with its interaction time. In email reading mode, each piece of email is considered to be a document and is added to the buffer. The type of interaction with the email item, such as deleting, storing, or forwarding, the sender of the email, and the recipient list are recorded. In email writing mode, each piece of written email is considered a document and added to the buffer. The recipient of the email is recorded. Documents written during document writing mode are added to the buffer. The user's access time with each piece of pushed information and type of interaction, such as saving or forwarding, are recorded. In finding expert advice mode, the user's interest in expert advice is recorded; interest may be measured by the interaction time with an email from an expert, a user's direct rating of the quality of information received, or other suitable measure.

During a product purchasing mode, a similar buffer is created for purchased products, as shown in FIG. 16. All purchased products are used to update the User Model. The user recently purchased products buffer records for each purchase the product ID, parent node in the product tree, purchase time, and purchase source. Purchased products are used to update the user product distribution and user product feature distribution.

If the user feels that the User Model is not an adequate representation of him or her, the user may submit user modification requests. For example, the user may request that specific web sites, topics, or phrases be added to or deleted from the User Model.

User Models for prototype users (hats) are also updated based on actions of similar users. Obviously, it is desirable for prototype User Models to reflect the current state of the representative interest. New web sites appear constantly, and even new informative words appear regularly. For example, technology-related words are introduced and widely adopted quite rapidly; the word list of the Technologist hat should be updated to reflect such changes.

Prototype User Models are updated using actions that are related to the prototype. Actions include documents, user reactions to documents, and product purchases. There are many ways to determine whether an action is relevant to the prototype user. A document that is a positive example for many users (i.e., a followed search result or bookmarked page) and also has a high probability of interest to the prototype user is added to the set of prototype user documents. Actions of users or clusters who are similar to the prototype user, as measured by the relative entropy between individual distributions (words, sites, etc.), are incorporated into the prototype User Model. Additions to the prototype User Model may be weighted by the relative entropy between the user performing the action and the prototype user. Actions of expert users who have a high degree of interest in topics also of interest to the prototype user are incorporated into the prototype User Model.

Note that users who are trying on hats are not able to change the prototype User Model. Their actions affect their own User Models, but not the prototype User Model. Updates to the prototype User Model are based only on actions of users who are not currently trying on hats.

Product models are also continually updated using incremental learning techniques. As described below, the present invention includes crawling network documents and evaluating each document against User Models. Crawled documents are also evaluated by product models. Documents that are relevant to a particular product, as determined by the computed probability P(p|d), are used to update its product model. If a document is determined to be relevant, then each component of the product model is updated accordingly. In addition to the parsing and analysis performed for user documents, information extraction techniques are employed to derive feature values that are compared against feature values of the product model, and also incorporated into the feature value list as necessary. New products can be added to the product tree at any time, with characteristic product feature values extracted from all relevant documents. Relevant documents for updating product models include product releases, discussion group entries, product reviews, news articles, or any other type of document.

By employing dynamically updated product models, the present invention, in contrast with prior art systems, provides for deep analysis of all available product information to create a rich representation of products. The interest of a user in a product can therefore be determined even if the product has never been purchased before, or if the product has only been purchased by a very small number of users.

Applying the User Model to Unseen Documents

The User Model is applied to unseen documents to determine the probability that a document is of interest to the user, or the probability that a document is of interest to a user in a particular context. The basic functionality of this determination is then used in the various applications described in subsequent sections to provide personalized information and product services to the user.

The process of estimating user interest in a particular unseen document 120 is illustrated in FIG. 17. This process has the following three steps:

  • 1. Preprocessing the document as for initialization (step 122).
  • 2. Calculating an individual score for the document for each element of the user representation (e.g., topic distribution, word list).
  • 3. Non-linearly combining (124) individual scores into one score 126, the probability that the user is interested in the unseen document, P(u|d).

The second step varies for each individual score. From the parsed text, the words of the document 120 are intersected with the words or phrases in the user informative word list 128. For every word or phrase in common, the stored mutual information between the two indicator variables Iw and Iu is summed to obtain the word score. Alternatively, the TFIDF associated with the word are averaged for every common word or phrase. The location score is given by the probability that the document site is of interest to the user, based on the user site distribution 130.

The topic classifiers 132 are applied to document 120 to determine the probability that the document relates to a particular topic, P(t|d). The user topic score is obtained by computing the relative entropy between the topic distribution P(t|d) and the user topic distribution 134, P(t|u). After the document has been classified into topics, the topic expert models 136 are applied as described above to determine a score reflecting the interest of users that are experts in the particular topics of this document.

Similarly, the product models 138 are applied to document 120 to determine which products or product categories it describes, P(p|d). From the document product distribution, the product score is obtained by computing the relative entropy between the document product distribution and user product distribution 140, P(p|u). For each product having a nonzero value of P(p|d), its feature values are given by the product model. The user's measures on each of these feature values, found in the user product feature distribution 141, are averaged to obtain a product feature score for each relevant product. Product feature scores are then averaged to obtain an overall product feature score.

The cluster models 142 of clusters to which the user belongs are applied to the document to obtain P(c(u)|d). This group model represents the average interests of all users in the cluster. Conceptually, the cluster model is obtained from the union of all the member users' documents and product purchases. Practically, the cluster model is computed from the User Models by averaging the different distributions of the individual User Models, and not from the documents or purchases themselves. Note that in a recursive way, all users have some impact (relative to their similarity to the user under discussion) on the user score, given that P(c(u)|d)) is estimated using P(c(c(u))|d) as a knowledge source, and so on.

Finally, world knowledge (not shown) is an additional knowledge source that represents the interest of an average user in the document based only on a set of predefined factors. World knowledge factors include facts or knowledge about the document, such as links pointing to and from the document or metadata about the document, for example, its author, publisher, time of publication, age, or language. Also included may be the number of users who have accessed the document, saved it in a favorites list, or been previously interested in the document. World knowledge is represented as a probability between 0 and 1.

In step 124, all individual scores are combined to obtain a composite user score 126 for document 120. Step 124 may be performed by training a Multilayer Perceptron using jackknifing and cross-validation techniques, as described in H. Bourlard and N. Morgan, Connectionist Speech Recognition: A Hybrid Approach, Kluwer Academic Publishers, 1994. It has been shown in J. Hertz et al., Introduction to The Theory of Neural Computation”, Addison-Wesley, 1991, that a Multilayer Perceptron can be trained to estimate posterior probabilities.

The context of a user's interaction can be explicitly represented in calculating the user interest in a document. It is not feasible to update the user model after every newly viewed document or search, but the User Model can be updated effectively instantaneously by incorporating the context of user interactions. Context includes content and location of documents viewed during the current interaction session. For example, if the user visits ten consecutive sites pertaining to computer security, then when the User Model estimates the interest of the user in a document about computer security, it is higher than average. The probability of user interest in a document within the current context con is given by: P ( u d , con ) = P ( u , con d ) P ( con d )

In some applications, individual scores that are combined in step 124 are themselves useful. In particular, the probability that a user is interested in a given product can be used to suggest product purchases to a user. If a user has previously purchased a product, then the User Model contains a distribution on the product's features. If these features propagate far up the product tree, then they can be used to estimate the probability that the user is interested in a different type of product characterized by similar features. For example, if the user purchases a digital camera that is Windows compatible, then the high probability of this compatibility feature value propagates up the tree to a higher node. Clearly, all computer-related purchases for this user should be Windows compatible. Every product that is a descendent of the node to which the value propagated can be rated based on its compatibility, and Windows-compatible products have a higher probability of being of interest to the user.

The long-term interest of a user in products, represented by P(p|u), is distinct from the user's immediate interest in a product p, represented as P(u|d, product described=p). The user's immediate interest is the value used to recommend products to a user. Note that P(p|u) does not incorporate the user's distribution on feature values. For example, consider the problem of evaluating a user's interest in a particular camera, the Nikon 320. The user has never read any documents describing the Nikon 320, and so P(Nikon 320|u)=0. However, the user's feature distribution for the Cameras node indicates high user interest in all of the feature values characterizing the Nikon 320.

When a given product is evaluated by the User Model, the following measures are combined to obtain P(u|d, product described=p): the probabilities of the product and its ancestor nodes from the user product distribution, P(p|u); an average of probabilities of each feature value from the user product feature distribution, P(f|u,p); a probability from the user's clusters' product distributions, P(f|c(u),p); and an average of probabilities of feature values from the cluster' product feature distributions, P(f|c(u),p). The overall product score is determined by non-linearly combining all measures. The cluster model is particularly useful if the user does not have a feature value distribution on products in which the user's interest is being estimated.

Applications

The basic function of estimating the probability that a user is interested in a document or product is exploited to provide different types of personalized services to the user. In each type of service, the user's response to the service provided is monitored to obtain positive and negative examples that are used to update the User Model. Example applications are detailed below. However, it is to be understood that all applications employing a trainable User Model as described above are within the scope of the present invention.

Personal Search

In this application, both the collection and filtering steps of searching are personalized. A set of documents of interest to the user is collected, and then used as part of the domain for subsequent searches. The collected documents may also be used as part of the user documents to update the User Model. The collection step, referred to as Personal Crawler, is illustrated schematically in FIG. 18. A stack 170 is initialized with documents of high interest to the user, such as documents in the bookmarks file or documents specified by the user. If necessary, the stack documents may be selected by rating each document in the general document index according to the User Model. The term “stack” refers to a pushdown stack as described in detail in R. Sedgewick, Algorithms in C++, Parts 1–4, Addison-Wesley, 1998.

In step 172, the crawler selects a document from the top of the stack to begin crawling. The document is parsed and analyzed (step 174) to identify any links to other documents. If there are links to other documents, each linked document is scored using the User Model (176). If the linked document is of interest to the user (178), i.e., if P(u|d) exceeds a threshold level, then it is added to the stack in step 180, and the crawler continues crawling from the linked document (step 172). If the document is not of interest to the user, then the crawler selects the next document on the stack to continue crawling.

The subsequent searching step is illustrated in FIG. 19. In response to a query 190, a set of search results is located from the set containing all documents D and user documents obtained during personal crawling. The results are evaluated using the User Model (194) and sorted in order of user interest (196), so that the most interesting documents are listed first. The user reaction to each document in the search results is monitored. Monitored reactions include whether or not a document was viewed or ignored and the time spent viewing the document. Documents to which the user responds positively are parsed and analyzed (200) and then used to update the User Model (202) as described above.

The role of the User Model in filtering the search results in step 194 is based on Bayesian statistics and pattern classification theory. According to pattern classification theory, as detailed in R. Duda and P. Hart, Pattern Classification and Scene Analysis, Wiley, 1973, the optimal search result is the one with the highest posterior probability. That is, the optimal result is given by: Max D P ( u q , d ) ,
where P(u|q,d) is the posterior probability of the event that a document d is of interest to a user u having an information need q. This probability can be expressed as: P ( u q , d ) = P ( q d , u ) P ( u d ) P ( q d ) .

The term P(u|d) represents the user interest in the document regardless of the current information need, and is calculated using the User Model. The term P(q|d,u) represents the probability that a user u with an information need of d expresses it in the form of a query q. The term P(q|d) represents the probability that an average user with an information need of d expresses it in the form of a query q. One possible implementation of the latter two terms uses the Hidden Markov Model, described in Christopher D. Manning and Hinrich Schutze, Foundations of Statistical Natural Language Processing, MIT Press, 1999.

Search results may also be filtered taking into account the context of user interactions, such as content of a recently viewed page or pages. When the context is included, the relevant equation is: P ( u q , d , con ) = P ( q d , u , con ) P ( u d , con ) P ( q d , con ) ,
where P(u|d,con) is as described above.

The Personal Crawler is also used to collect and index documents for product models. Collected documents are parsed and analyzed to update product models, particularly the list of product feature values, which are extracted from collected documents using information extraction techniques.

In general, searches are performed to retrieve all documents from the set of indexed documents that match the search query. Alternatively, searches can be limited to product-related documents, based on either the user's request, the particular search query, or the user's context. For example, a user is interested in purchasing a new bicycle. In one embodiment, the user selects a check-box or other graphical device to indicate that only product-related documents should be retrieved. When the box is not checked, a search query “bicycle” returns sites of bicycle clubs and newsletters. When the box is checked, only documents that have a nonzero product probability (P(p|d)) on specific products are returned. Such documents include product pages from web sites of bicycle manufacturers, product reviews, and discussion group entries evaluating specific bicycle models.

Alternatively, the search query itself is used to determine the type of pages to return. For example, a query “bicycle” again returns sites of bicycle clubs and newsletters. However, a query “cannondale bicycle” or “cannondale” returns only product-related pages for Cannondale bicycles. Alternatively, the user's context is used to determine the type of pages to return. If the last ten pages viewed by the user are product-related pages discussing Cannondale bicycles, then the query “bicycle” returns product-related pages for all brands of bicycles that are of interest to the user, as determined by the User Model. In all three possible embodiments, within the allowable subset of documents, the entire document is evaluated by the User Model to estimate the probability that the user is interested in the document.

Searches may also be performed for products directly, and not for product-related documents. Results are evaluated using only the user product distribution, user product feature distribution, and product and feature distributions of the user's clusters, as explained above. In general, product searches are performed only at the request of the user, for example by selecting a “product search” tab using a mouse or other input device. A user enters a product category and particular feature values, and a list of products that are estimated to be of high interest to the user is returned. The user is returned some form of list of most interesting products. The list may contain only the product name, and may include descriptions, links to relevant documents, images, or any other appropriate information.

Personal Browsing and Navigation

The present invention personalizes browsing and navigation in a variety of different ways. In the personal web sites application, web sites located on third party servers are written in a script language that enables dynamic tailoring of the site to the user interests. Parameters of the User Model are transferred to the site when a user requests a particular page, and only selected content or links are displayed to the user. In one embodiment, the site has different content possibilities, and each possibility is evaluated by the User Model. For example, the CNN home page includes several potential lead articles, and only the one that is most interesting to the user is displayed. In a second embodiment, links on a page are shown only if the page to which they link is of interest to the user. For example, following the lead article on the CNN home page are links to related articles, and only those of interest to the user are shown or highlighted. One single article has a variety of potential related articles; a story on the Microsoft trial, for example, has related articles exploring legal, technical, and financial ramifications, and only those meeting the user's information needs are displayed.

The personal links application is illustrated in FIG. 20. In this application, the hyperlinks in a document being viewed by the user are graphically altered, e.g., in their color, to indicate the degree of interest of the linked documents to the use. As a user views a document (step 210), the document is parsed and analyzed (212) to locate hyperlinks to other documents. The linked documents are located in step 214 (but not shown to the user), and evaluated with the User Model (214) to estimate the user's interest in each of the linked documents. In step 216, the graphical representation of the linked documents is altered in accordance with the score computed with the User Model. For example, the links may be color coded, with red links being most interesting and blue links being least interesting, changed in size, with large links being most interesting, or changed in transparency, with uninteresting links being faded. If the user follows one of the interesting links (218), then the process is repeated for the newly viewed document (210).

The personal related pages application locates pages related to a viewed page. Upon the user's request (e.g., by clicking a button with a mouse pointer), the related pages are displayed. Related pages are selected from the set of user documents collected by the personal crawler. Implementation is similar to that of the personal search application, with the viewed page serving as the query. Thus the relevant equation becomes P ( u page , d ) = P ( page d , u ) P ( u d ) P ( page d ) ,
with P(page|d,u) representing the probability that a user u with an information need of document d expresses it in the form of the viewed page page. P(page|d) represents the probability that an average user with an information need of document d expresses it in the form of the viewed page page. These terms can be calculated using the Hidden Markov Model.

Alternatively, related pages or sites may be selected according to the cluster model of clusters to which the user belongs. The most likely site navigation from the viewed site, based on the behavior of the cluster members, is displayed to user upon request.

Related pages are particularly useful in satisfying product information needs. For example, if the user is viewing a product page of a specific printer on the manufacturer's web site, clicking the “related pages” button returns pages comparing this printer to other printers, relevant newsgroup discussions, or pages of comparable printers of different manufacturers. All returned related pages have been evaluated by the User Model to be of interest to the user.

Find the Experts

In this application, expert users are located who meet a particular information or product need of the user. Expert users are users whose User Model indicates a high degree of interest in the information need of the user. The information need is expressed as a document or product that the user identifies as representing his or her need. In this context, a document may be a full document, a document excerpt, including paragraphs, phrases, or words, the top result of a search based on a user query, or an email message requesting help with a particular subject. From the pool of potential experts, User Models are applied to the document or product, and users whose probability of interest in the document or product exceeds a threshold level are considered expert users.

The pool of experts is specified either by the user or in the system. For example, the pool may include all company employees or users who have previously agreed to help and advise other users. When users request expert advice about a particular product, the expert may be chosen from the product manufacturer or from users who have previously purchased the product, or from users participating in discussion groups about the product.

A protocol for linking users and identified experts is determined. For example, the expert receives an email message requesting that he or she contact the user in need of assistance. Alternatively, all user needs are organized in a taxonomy of advice topics, and an expert searches for requests associated with his or her topic of expertise.

Personal News

This application, also known as personal pushed information, uses the personal crawler illustrated in FIG. 18. From all documents collected within a recent time period by the user's crawler or user's clusters' crawlers, the most interesting ones are chosen according to the User Model. Collection sources may also be documents obtained from news wires of actions of other users. Documents are sent to the user in any suitable manner. For example, users receive email messages containing URLs of interesting pages, or links are displayed on a personal web page that the user visits.

Personalization Assistant

Using the User Model, the Personalization Assistant can transform any services available on the web into personalized services, such as shopping assistants, chatting browsers, or matchmaking assistants.

Document Barometer

The document barometer, or Page-O-Meter, application, illustrated in FIG. 21, finds the average interest of a large group of users in a document. The barometer can be used by third parties, such as marketing or public relations groups, to analyze the interest of user groups in sets of documents, advertising, or sites, and then modify the documents or target advertising at particular user groups. The application can instead report a score for a single user's interest in a document, allowing the user to determine whether the system is properly evaluating his or her interest. If not, the user can make user modification requests for individual elements of the User Model. From individual and average scores, the application determines a specific user or users interested in the document.

Referring to FIG. 21, a document 220 is parsed and analyzed (222) and then evaluated according to a set of N User Models 224 and 226 through 228. N includes any number greater than or equal to one. The resulting scores from all User Models are combined and analyzed in step 230. In one embodiment, the analysis locates users having maximum interest in document 220, or interest above a threshold level, and returns a sorted list of interested users (232). Alternatively, an average score for document 220 is calculated and returned (234). The average score may be for all users or for users whose interest exceeds a threshold interest level. The range of interest levels among all users in the group may also be reported.

An analogous product barometer calculates user interest in a product. The product barometer computes a score for an individual user or group of users, or identifies users having an interest in a product that exceeds a threshold level. Third party organizations user the product barometer to target marketing efforts to users who are highly likely to be interested in particular products.

3D Map

FIG. 22 illustrates a three-dimensional (3D) map 240 of the present invention, in which rectangles represent documents and lines represent hyperlinks between documents. A user provides a set of hyperlinked documents, and each document is scored according to the User Model. An image of 3D map 240 is returned to the user. 3D map 240 contains, for each document, a score reflecting the probability of interest of the user in the document.

Product Recommendations

A user's online shopping experience can be personalized by making use of the user's overall product score described above, P(u|d, product described=p). Products that are of high interest to the user are suggested to him or her for purchase. When a user requests information for a specific product or purchases a product, related products are suggested (up-sell). Related product categories are predetermined by a human, but individual products within related categories are evaluated by the User Model before being suggested to the user. The related products are given to the user in a list that may contain images, hyperlinks to documents, or any other suitable information. For example, when a user purchases a server, a list of relevant backup tapes are suggested to him or her for purchase. Suggested products may have feature values that are known to be of interest to the user, or may have been purchased by other members of the user's cluster who also purchased the server. Related product suggestions may be made at any time, not only when a user purchases or requests information about a particular product. Suggested products may be related to any previously purchased products.

Similarly, competing or comparable products are suggested to the user (cross-sell). When the user browses pages of a particular product, or begins to purchase a product, products within the same product category are evaluated to estimate the user's interest in them. Products that are highly interesting to the user are recommended. The user might intend to purchase one product, but be shown products that are more useful or interesting to him or her.

It will be clear to one skilled in the art that the above embodiments may be altered in many ways without departing from the scope of the invention. Accordingly, the scope of the invention should be determined by the following claims and their legal equivalents.

Claims (62)

1. A computer-implemented method for providing automatic, personalized information services to a user u, the method comprising:
a) transparently monitoring user interactions with data while the user is engaged in normal use of a computer;
b) updating user-specific data files, wherein the user-specific data files comprise the monitored user interactions with the data and a set of documents associated with the user;
c) estimating parameters of a learning machine, wherein the parameters define a User Model specific to the user and wherein the parameters are estimated in part from the user-specific data files;
d) analyzing a document d to identify properties of the document;
e) estimating a probability P(u|d) that an unseen document d is of interest to the user u, wherein the probability P(u|d) is estimated by applying the identified properties of the document to the learning machine having the parameters defined by the User Model; and
f) using the estimated probability to provide automatic, personalized information services to the user.
2. The method of claim 1 wherein the user-specific data files include documents of interest to the user u and documents that are not of interest to the user u, and wherein estimating the parameters comprises distinct treatment of the documents of interest and the documents that are not of interest.
3. The method of claim 1 wherein analyzing the document d provides for the analysis of documents having multiple distinct media types.
4. The method of claim 1 wherein transparently monitoring user interactions with data comprises monitoring multiple distinct modes of user interaction with network data.
5. The method of claim 4 wherein the multiple distinct modes of user interaction comprise a mode selected from the group consisting of a network searching mode, a network navigation mode, a network browsing mode, an email reading mode, an email writing mode, a document writing mode, a viewing “pushed” information mode, a finding expert advice mode, and a product purchasing mode.
6. The method of claim 1 further comprising crawling network documents, wherein the crawling comprises parsing crawled documents for links, calculating probable user interest in the parsed links using the learning machine, and preferentially following links likely to be of interest to the user.
7. The method of claim 1 wherein the identified properties of the document d comprise a user u-independent property selected from the group consisting of:
a) a probability P(t,d) that the document d is of interest to users interested in a topic t;
b) a topic classifier discrete probability distribution P(t|d);
c) a product model discrete probability distribution P(p|d);
d) product feature values extracted from the document d;
e) an author of the document d;
f) an age of the document d;
g) a list of documents linked to the document d;
h) a language of the document d;
i) a number of users who have accessed the document d;
j) a number of users who have saved the document d in a favorite document list; and
k) a list of users previously interested in the document d.
8. The method of claim 1 wherein the parameters of the learning machine define a user u-dependent function selected from the group consisting of:
a) a user topic probability distribution P(t|u) representing interests of the user u in various topics t;
b) a user product probability distribution P(p|u) representing interests of the user u in various products p;
c) a user product feature probability distribution P(f|u,p) representing interests of the user u in various features f of each of the various products p;
d) a web site probability distribution P(s|u) representing interests of the user u in various web sites s;
e) a cluster probability distribution P(c(u)|u) representing similarity of the user u to users in various clusters c(u);
f) a phrase model probability distribution P(w|u) representing interests of the user u in various phrases w;
g) an information theory based measure I(Iw; Iu) representing mutual information between various phrases w and the user u;
h) an information theory based measure I(It; Iu) representing mutual information between various topics t and the user u;
i) an information theory based measure I(Is; Iu) representing mutual information between various web sites s and the user u;
j) an information theory based measure I(Ip; Iu) representing mutual information between various products p and the user u; and
k) an information theory based measure I(If; Iu) representing mutual information between various features f of each of the various products p and the user u.
9. The method of claim 1 wherein the parameters of the learning machine define:
a) a user product probability distribution P(p|u) representing interests of the user u in various products p; and
b) a user product feature probability distribution P(f|u,p) representing interests of the user u in various features f of each of the various products p; and wherein the method further comprises estimating a probability P(u|d, product described=p) that a document d that describes a product p is of interest to the user u, wherein the probability is estimated in part from the user product probability distribution and the user product feature probability distribution.
10. The method of claim 9 further comprising recommending products to the user based on the probability P(u|d, product described=p).
11. The method of claim 1 further comprising estimating a posterior probability P(u|d,q) that the document d is of interest to the user u, given a query q submitted by the user.
12. The method of claim 11 wherein estimating the posterior probability comprises estimating a probability P(q|d,u) that the query q is expressed by the user u with an information need in the document d.
13. The method of claim 1 further comprising applying the identified properties of the document d to a learning machine having product parameters characterizing a product p to estimate a probability P(p|d) that the document d refers to the product p.
14. The method of claim 13 further comprising updating the product parameters based on the identified properties of the document d and the estimated probability P(p|d).
15. The method of claim 13 further comprising initializing the product parameters based on a set of documents associated with the product p.
16. The method of claim 1 further comprising clustering multiple users into clusters of similar users, wherein the clustering comprises calculating distances between User Models, and selecting similar users based on the calculated distances between User Models.
17. The method of claim 1 further comprising calculating relative entropy values between User Models of multiple users, and clustering together users based on the calculated relative entropy values.
18. The method of claim 1 wherein the parameters defining the User Model comprise calculated distances between the User Model and User Models of users similar to the user.
19. The method of claim 1 further comprising selecting in a group of users an expert user in an area of expertise, wherein selecting the expert user comprises finding an expert User Model among User Models of the group of users, such that the expert User Model indicates a strong interest of the expert user in a document associated with the area of expertise.
20. The method of claim 1 further comprising parsing the document d for hyperlinks, and separately estimating for each of the hyperlinks a probability that the hyperlink is of interest to the user u.
21. The method of claim 1 further comprising sending to a third party web server user interest information derived from the User Model, whereby the third party web server may customize its interaction with the user.
22. The method of claim 1 wherein the monitored user interactions include a sequence of interaction times.
23. The method of claim 1 further comprising initializing the User Model using information selected from the group consisting of a set of documents provided by the user, a web browser history file associated with the user, a web browser bookmarks file associated with the user, ratings by the user of a set of documents, and previous product purchases made by the user.
24. The method of claim 1 further comprising modifying the User Model based on User Model modification requests provided by the user.
25. The method of claim 1 further comprising providing to the user a score for a document identified by the user, wherein the score is derived from the estimated probability.
26. The method of claim 1 further comprising providing to the user a 3D map of a hyper linked document collection, wherein the 3D map indicates a user interest in each document.
27. The method of claim 1 further comprising temporarily using a User Model that is built from a set of predetermined parameters of a profile selected by the user.
28. The method of claim 1 further comprising initializing the User Model by selecting a set of predetermined parameters of a prototype user selected by the user.
29. The method of claim 28 further comprising updating the predetermined parameters of the prototype user based on actions of users similar to the prototype user.
30. The method of claim 1 further comprising identifying a set of users interested in the document d.
31. The method of claim 30 further comprising calculating a range of interests in the document d for the identified set of users.
32. A program storage device accessible by a central computer, tangibly embodying a program of instructions executable by the central computer to perform method steps for providing automatic, personalized information services to a user u, the method steps comprising:
a) transparently monitoring user interactions with data while the user is engaged in normal use of a client computer in communication with the central computer;
b) updating user-specific data files, wherein the user-specific data files comprise the monitored user interactions with the data and a set of documents associated with the user;
c) estimating parameters of a learning machine, wherein the parameters define a User Model specific to the user and wherein the parameters are estimated in part from the user-specific data files;
d) analyzing a document d to identify properties of the document;
e) estimating a probability P(u|d) that an unseen document d is of interest to the user u, wherein the probability P(u|d) is estimated by applying the identified properties of the document to the learning machine having the parameters defined by the User Model; and
f) using the estimated probability to provide automatic, personalized information services to the user.
33. The program storage device of claim 32 wherein the user-specific data files include documents of interest to the user u and documents that are not of interest to the user u, and wherein estimating the parameters comprises distinct treatment of the documents of interest and the documents that are not of interest.
34. The program storage device of claim 32 wherein analyzing the document d provides for the analysis of documents having multiple distinct media types.
35. The program storage device of claim 32 wherein transparently monitoring user interactions with data comprises monitoring multiple distinct modes of user interaction with network data.
36. The program storage device of claim 35 wherein the multiple distinct modes of user interaction comprise a mode selected from the group consisting of a network searching mode, a network navigation mode, a network browsing mode, an email reading mode, an email writing mode, a document writing mode, a viewing “pushed” information mode, a finding expert advice mode, and a product purchasing mode.
37. The program storage device of claim 32 wherein the method steps further comprise crawling network documents, wherein the crawling comprises parsing crawled documents for links, calculating probable user interest in the parsed links using the learning machine, and preferentially following links likely to be of interest to the user.
38. The program storage device of claim 32 wherein the identified properties of the document d comprise a user u-independent property selected from the group consisting of:
a) a probability P(t,d) that the document d is of interest to users interested in a topic t;
b) a topic classifier discrete probability distribution P(t|d);
c) a product model discrete probability distribution P(p|d);
d) product feature values extracted from the document d;
e) an author of the document d;
f) an age of the document d;
g) a list of documents linked to the document d;
h) a language of the document d;
i) a number of users who have accessed the document d;
j) a number of users who have saved the document d in a favorite document list; and
k) a list of users previously interested in the document d.
39. The program storage device of claim 32 wherein the parameters of the learning machine define a user u-dependent function selected from the group consisting of:
a) a user topic probability distribution P(t|u) representing interests of the user u in various topics t;
b) a user product probability distribution P(p|u) representing interests of the user u in various products p;
c) a user product feature probability distribution P(f|u,p) representing interests of the user u in various features f of each of the various products p;
d) a web site probability distribution P(s|u) representing interests of the user u in various web sites s;
e) a cluster probability distribution P(c(u)|u) representing similarity of the user u to users in various clusters c(u);
f) a phrase model probability distribution P(w|u) representing interests of the user u in various phrases w;
g) an information theory based measure I(Iw; Iu) representing mutual information between various phrases w and the user u;
h) an information theory based measure I(It; Iu) representing mutual information between various topics t and the user u;
i) an information theory based measure I(Is; Iu) representing mutual information between various web sites s and the user u;
j) an information theory based measure I(Ip; Iu) representing mutual information between various products p and the user u; and
k) an information theory based measure I(If; Iu) representing mutual information between various features f of each of the various products p and the user u.
40. The program storage device of claim 32 wherein the parameters of the learning machine define:
a) a user product probability distribution P(p|u) representing interests of the user u in various products p; and
b) a user product feature probability distribution P(f|u,p) representing interests of the user u in various features f of each of the various products p;
and wherein the method steps further comprise estimating a probability P(u|d, product described=p) that a document d that describes a product p is of interest to the user u, wherein the probability is estimated in part the user product probability distribution and the user product feature probability distribution.
41. The program storage device of claim 40 wherein the method steps further comprise recommending products to the user based on the probability P(u|d, product described=p).
42. The program storage device of claim 32 wherein the method steps further comprise estimating a posterior probability P(u|d,q) that the document d is of interest to the user u, given a query q submitted by the user.
43. The program storage device of claim 42 wherein estimating the posterior probability comprises estimating a probability P(q|d,u) that the query q is expressed by the user u with an information need in the document d.
44. The program storage device of claim 32 wherein the method steps further comprise applying the identified properties of the document d to a learning machine having product parameters characterizing a product p to estimate a probability P(p|d) that the document d refers to the product p.
45. The program storage device of claim 44 wherein the method steps further comprise updating the product parameters based on the identified properties of the document d and the estimated probability P(p|d).
46. The program storage device of claim 44 wherein the method steps further comprise initializing the product parameters based on a set of documents associated with the product p.
47. The program storage device of claim 32 wherein the method steps further comprise clustering multiple users into clusters of similar users, wherein the clustering comprises calculating distances between User Models, and selecting similar users based on the calculated distances between User Models.
48. The program storage device of claim 32 wherein the method steps further comprise calculating relative entropy values between User Models of multiple users, and clustering together users based on the calculated relative entropy values.
49. The program storage device of claim 32 wherein the parameters defining the User Model comprise calculated distances between the User Model and User Models of users similar to the user.
50. The program storage device of claim 32 wherein the method steps further comprise selecting in a group of users an expert user in an area of expertise, wherein selecting the expert user comprises finding an expert User Model among User Models of the group of users, such that the expert User Model indicates a strong interest of the expert user in a document associated with the area of expertise.
51. The program storage device of claim 32 wherein the method steps further comprise parsing the document d for hyperlinks, and separately estimating for each of the hyperlinks a probability that the hyperlink is of interest to the user u.
52. The program storage device of claim 32 wherein the method steps further comprise sending to a third party web server user interest information derived from the User Model, whereby the third party web server may customize its interaction with the user.
53. The program storage device of claim 32 wherein the monitored user interactions include a sequence of interaction times.
54. The program storage device of claim 32 wherein the method steps further comprise initializing the User Model using information selected from the group consisting of a set of documents provided by the user, a web browser history file associated with the user, a web browser bookmarks file associated with the user, ratings by the user of a set of documents, and previous product purchases made by the user.
55. The program storage device of claim 32 wherein the method steps further comprise modifying the User Model based on User Model modification requests provided by the user.
56. The program storage device of claim 32 wherein the method steps further comprise providing to the user a score for a document identified by the user, wherein the score is derived from the estimated probability.
57. The program storage device of claim 32 wherein the method steps further comprise providing to the user a 3D map of a hyper linked document collection, wherein the 3D map indicates a user interest in each document.
58. The program storage device of claim 32 wherein the method steps further comprise temporarily using a User Model that is built from a set of predetermined parameters of a profile selected by the user.
59. The program storage device of claim 32 wherein the method steps further comprise initializing the User Model by selecting a set of predetermined parameters of a prototype user selected by the user.
60. The program storage device of claim 59 wherein the method steps further comprise updating the predetermined parameters of the prototype user based on actions of users similar to the prototype user.
61. The program storage device of claim 32 wherein the method steps further comprise identifying a set of users interested in the document d.
62. The program storage device of claim 61 wherein the method steps further comprise calculating a range of interests in the document d for the identified set of users.
US09597975 1999-12-28 2000-06-20 Automatic, personalized online information and product services Active 2022-12-21 US6981040B1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US17339299 true 1999-12-28 1999-12-28
US09597975 US6981040B1 (en) 1999-12-28 2000-06-20 Automatic, personalized online information and product services

Applications Claiming Priority (7)

Application Number Priority Date Filing Date Title
US09597975 US6981040B1 (en) 1999-12-28 2000-06-20 Automatic, personalized online information and product services
US11316785 US7320031B2 (en) 1999-12-28 2005-12-22 Automatic, personalized online information and product services
US12008148 US7685276B2 (en) 1999-12-28 2008-01-08 Automatic, personalized online information and product services
US12692252 US8301764B2 (en) 1999-12-28 2010-01-22 Method and system for personalized searching of information and product services by estimating an interest to a user
US13619427 US8626911B2 (en) 1999-12-28 2012-09-14 Automatic, personalized online information and product services
US13907115 US8825853B1 (en) 1999-12-28 2013-05-31 Automatic, personalized online information and product services
US14178715 US9098813B1 (en) 1999-12-28 2014-02-12 Methods and systems for conducting a personalized web browser search by probability estimation that is unseen document is of interest to the user

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US11316785 Continuation US7320031B2 (en) 1999-12-28 2005-12-22 Automatic, personalized online information and product services

Publications (1)

Publication Number Publication Date
US6981040B1 true US6981040B1 (en) 2005-12-27

Family

ID=35482767

Family Applications (7)

Application Number Title Priority Date Filing Date
US09597975 Active 2022-12-21 US6981040B1 (en) 1999-12-28 2000-06-20 Automatic, personalized online information and product services
US11316785 Active US7320031B2 (en) 1999-12-28 2005-12-22 Automatic, personalized online information and product services
US12008148 Active US7685276B2 (en) 1999-12-28 2008-01-08 Automatic, personalized online information and product services
US12692252 Active 2020-09-02 US8301764B2 (en) 1999-12-28 2010-01-22 Method and system for personalized searching of information and product services by estimating an interest to a user
US13619427 Active US8626911B2 (en) 1999-12-28 2012-09-14 Automatic, personalized online information and product services
US13907115 Active US8825853B1 (en) 1999-12-28 2013-05-31 Automatic, personalized online information and product services
US14178715 Active US9098813B1 (en) 1999-12-28 2014-02-12 Methods and systems for conducting a personalized web browser search by probability estimation that is unseen document is of interest to the user

Family Applications After (6)

Application Number Title Priority Date Filing Date
US11316785 Active US7320031B2 (en) 1999-12-28 2005-12-22 Automatic, personalized online information and product services
US12008148 Active US7685276B2 (en) 1999-12-28 2008-01-08 Automatic, personalized online information and product services
US12692252 Active 2020-09-02 US8301764B2 (en) 1999-12-28 2010-01-22 Method and system for personalized searching of information and product services by estimating an interest to a user
US13619427 Active US8626911B2 (en) 1999-12-28 2012-09-14 Automatic, personalized online information and product services
US13907115 Active US8825853B1 (en) 1999-12-28 2013-05-31 Automatic, personalized online information and product services
US14178715 Active US9098813B1 (en) 1999-12-28 2014-02-12 Methods and systems for conducting a personalized web browser search by probability estimation that is unseen document is of interest to the user

Country Status (1)

Country Link
US (7) US6981040B1 (en)

Cited By (318)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20020040386A1 (en) * 2000-08-31 2002-04-04 Bertrand Berthelot Method and device for adapting the content of documents of an information server
US20020069223A1 (en) * 2000-11-17 2002-06-06 Goodisman Aaron A. Methods and systems to link data
US20020198962A1 (en) * 2001-06-21 2002-12-26 Horn Frederic A. Method, system, and computer program product for distributing a stored URL and web document set
US20030014399A1 (en) * 2001-03-12 2003-01-16 Hansen Mark H. Method for organizing records of database search activity by topical relevance
US20030101083A1 (en) * 2001-10-30 2003-05-29 Sambedu Venkatesh Web-based method and system for providing expert information on selected matters
US20030101451A1 (en) * 2001-01-09 2003-05-29 Isaac Bentolila System, method, and software application for targeted advertising via behavioral model clustering, and preference programming based on behavioral model clusters
US20030172143A1 (en) * 2002-03-06 2003-09-11 Koji Wakayama Access node apparatus and method for internet using condition analysis
US20030171977A1 (en) * 2002-03-07 2003-09-11 Compete, Inc. Clickstream analysis methods and systems
US20030217097A1 (en) * 2002-03-21 2003-11-20 Eitel Robert T. Search agent for searching the internet
US20040143667A1 (en) * 2003-01-17 2004-07-22 Jason Jerome Content distribution system
US20040210541A1 (en) * 2001-05-04 2004-10-21 Jeremy Epstien User interface for a rules engine and methods therefor
US20040210558A1 (en) * 2001-11-19 2004-10-21 Fujitsu Limited Information navigation system
US20050004943A1 (en) * 2003-04-24 2005-01-06 Chang William I. Search engine and method with improved relevancy, scope, and timeliness
US20050010951A1 (en) * 2003-05-14 2005-01-13 Sony Corporation Information processing apparatus and method, program, and recording medium
US20050038638A1 (en) * 2003-07-25 2005-02-17 Dorin Comaniciu Density morphing and mode propagation for Bayesian filtering
US20050071328A1 (en) * 2003-09-30 2005-03-31 Lawrence Stephen R. Personalization of web search
US20050114382A1 (en) * 2003-11-26 2005-05-26 Lakshminarayan Choudur K. Method and system for data segmentation
US20050131866A1 (en) * 2003-12-03 2005-06-16 Badros Gregory J. Methods and systems for personalized network searching
US20050131762A1 (en) * 2003-12-31 2005-06-16 Krishna Bharat Generating user information for use in targeted advertising
US20050137886A1 (en) * 2003-12-17 2005-06-23 Kim Littrell Methods, systems and storage medium for providing content notification based on user communities
US20050144084A1 (en) * 2003-12-31 2005-06-30 Gold Kevin K. Using a customer's purchasing intent in recommending alternative items for purchase in a virtual store
US20050149498A1 (en) * 2003-12-31 2005-07-07 Stephen Lawrence Methods and systems for improving a search ranking using article information
US20050149619A1 (en) * 2003-12-22 2005-07-07 Cowan Keith O. Methods, systems and storage medium for distributing content based on user communities
US20050159204A1 (en) * 2002-02-14 2005-07-21 Martin Moshal Menu selection system and method of operation thereof
US20050192949A1 (en) * 2004-02-27 2005-09-01 Yuichi Kojima Document group analyzing apparatus, a document group analyzing method, a document group analyzing system, a program, and a recording medium
US20050228642A1 (en) * 2004-04-06 2005-10-13 Microsoft Corporation Efficient capitalization through user modeling
US20060015504A1 (en) * 2004-07-15 2006-01-19 Qingfeng Yu Method and system for site path evaluation using web session clustering
US20060026642A1 (en) * 2002-12-11 2006-02-02 Koninklijke Philips Electronics, N.V. Method and apparatus for predicting a number of individuals interested in an item based on recommendations of such item
US20060036728A1 (en) * 2004-06-18 2006-02-16 Fortinet, Inc. Systems and methods for categorizing network traffic content
US20060074883A1 (en) * 2004-10-05 2006-04-06 Microsoft Corporation Systems, methods, and interfaces for providing personalized search and information access
US20060106792A1 (en) * 2004-07-26 2006-05-18 Patterson Anna L Multiple index based information retrieval system
US20060136589A1 (en) * 1999-12-28 2006-06-22 Utopy, Inc. Automatic, personalized online information and product services
US20060195583A1 (en) * 2003-02-27 2006-08-31 Fabio Bellifemine Method and system for providing information services to a client using a user profile
US20060200433A1 (en) * 2004-11-04 2006-09-07 Manyworlds, Inc. Adaptive Self-Modifying and Recombinant Systems
US20060224554A1 (en) * 2005-03-29 2006-10-05 Bailey David R Query revision using known highly-ranked queries
US20060246932A1 (en) * 2001-01-16 2006-11-02 Texas Instruments Incorporated Collaborative Mechanism of Enhanced Coexistence of Collocated Wireless Networks
US20060271846A1 (en) * 2005-05-24 2006-11-30 Microsoft Corporation Systems and methods that facilitate improved display of electronic documents
US20060288017A1 (en) * 2005-06-17 2006-12-21 Goetz Oliver C User-specific product documentation
US20070005646A1 (en) * 2005-06-30 2007-01-04 Microsoft Corporation Analysis of topic dynamics of web search
US20070011203A1 (en) * 2005-05-31 2007-01-11 Tomohiro Tsunoda Information processing device, method of information processing, and program
US20070011131A1 (en) * 2002-05-17 2007-01-11 Delefevre Patrick Y Neutral sales consultant
US20070015118A1 (en) * 2005-07-14 2007-01-18 Red Hat, Inc. Tutorial generator with automatic capture of screenshots
US20070022082A1 (en) * 2005-07-20 2007-01-25 International Business Machines Corporation Search engine coverage
US20070027864A1 (en) * 2005-07-29 2007-02-01 Collins Robert J System and method for determining semantically related terms
US20070027865A1 (en) * 2005-07-29 2007-02-01 Kevin Bartz System and method for determining semantically related term
US20070033230A1 (en) * 2000-11-07 2007-02-08 Kiyohisa Minowa System, method, apparatus, and program for providing electronic manual
US20070038634A1 (en) * 2005-08-09 2007-02-15 Glover Eric J Method for targeting World Wide Web content and advertising to a user
US20070073553A1 (en) * 2004-05-20 2007-03-29 Manyworlds, Inc. Adaptive Commerce Systems and Methods
US20070083343A1 (en) * 2001-06-08 2007-04-12 Panoratio Database Images Gmbh Statistical models for improving the performance of database operations
US20070094157A1 (en) * 2005-04-08 2007-04-26 Manyworlds, Inc. Adaptive Digital Asset and Media Management
US20070112712A1 (en) * 2005-11-02 2007-05-17 Manyworlds, Inc. Adaptive Knowledge Lifecycle Management
US20070118546A1 (en) * 2005-11-21 2007-05-24 Sony Corporation User's preference prediction from collective rating data
US7231419B1 (en) * 2001-10-19 2007-06-12 Outlooksoft Corporation System and method for adaptively selecting and delivering recommendations to a requester
US20070156614A1 (en) * 2003-11-28 2007-07-05 Manyworlds, Inc. Adaptive Fuzzy Network System and Method
US20070162424A1 (en) * 2005-12-30 2007-07-12 Glen Jeh Method, system, and graphical user interface for alerting a computer user to new results for a prior search
US20070168342A1 (en) * 2005-12-29 2007-07-19 Singerman Brian A Recommended alerts
US20070203589A1 (en) * 2005-04-08 2007-08-30 Manyworlds, Inc. Adaptive Recombinant Process Methods
US20070208682A1 (en) * 2000-07-19 2007-09-06 Convergys Cmg Utah, Inc. Expert supported interactive product selection and recommendation
US20070250404A1 (en) * 2006-04-14 2007-10-25 At&T Corp. Automatic Learning For Mapping Spoken/Text Descriptions Of Products Onto Available Products
US20070255701A1 (en) * 2006-04-28 2007-11-01 Halla Jason M System and method for analyzing internet content and correlating to events
US20070288431A1 (en) * 2006-06-09 2007-12-13 Ebay Inc. System and method for application programming interfaces for keyword extraction and contextual advertisement generation
US20070288454A1 (en) * 2006-06-09 2007-12-13 Ebay Inc. System and method for keyword extraction and contextual advertisement generation
US20080010258A1 (en) * 2006-06-22 2008-01-10 Infosys Technologies Ltd. Collaborative filtering-based recommendations
US20080010281A1 (en) * 2006-06-22 2008-01-10 Yahoo! Inc. User-sensitive pagerank
US20080016101A1 (en) * 2003-12-30 2008-01-17 Shopping.Com Systems and methods for dynamically updating relevance of a selected item
US20080040313A1 (en) * 2006-08-11 2008-02-14 Yahoo! Inc. System and method for providing tag-based relevance recommendations of bookmarks in a bookmark and tag database
US20080040218A1 (en) * 2006-07-05 2008-02-14 Van Dijk Bob System and method for category-based contextual advertisement generation and management
US7333976B1 (en) 2004-03-31 2008-02-19 Google Inc. Methods and systems for processing contact information
US20080059419A1 (en) * 2004-03-31 2008-03-06 David Benjamin Auerbach Systems and methods for providing search results
US20080059395A1 (en) * 2006-01-10 2008-03-06 Manyworlds, Inc. Adaptive Online Experimentation
WO2008032037A1 (en) * 2006-09-11 2008-03-20 Stevens Richard J Method and system for filtering and searching data using word frequencies
US7353246B1 (en) * 1999-07-30 2008-04-01 Miva Direct, Inc. System and method for enabling information associations
US20080097821A1 (en) * 2006-10-24 2008-04-24 Microsoft Corporation Recommendations utilizing meta-data based pair-wise lift predictions
US20080104039A1 (en) * 2004-11-24 2008-05-01 Linda Lowson System and method for resource management
US20080114751A1 (en) * 2006-05-02 2008-05-15 Surf Canyon Incorporated Real time implicit user modeling for personalized search
US20080120292A1 (en) * 2006-11-20 2008-05-22 Neelakantan Sundaresan Search clustering
US20080120072A1 (en) * 2006-11-16 2008-05-22 Yahoo! Inc. System and method for determining semantically related terms based on sequences of search queries
US20080133573A1 (en) * 2004-12-24 2008-06-05 Michael Haft Relational Compressed Database Images (for Accelerated Querying of Databases)
US7389241B1 (en) * 2001-03-07 2008-06-17 Thomas Layne Bascom Method for users of a network to provide other users with access to link relationships between documents
US7392278B2 (en) * 2004-01-23 2008-06-24 Microsoft Corporation Building and using subwebs for focused search
US20080177778A1 (en) * 2002-03-07 2008-07-24 David Cancel Presentation of media segments
US20080183806A1 (en) * 2002-03-07 2008-07-31 David Cancel Presentation of media segments
US20080183805A1 (en) * 2002-03-07 2008-07-31 David Cancel Presentation of media segments
US20080183681A1 (en) * 2007-01-29 2008-07-31 Samsung Electronics Co., Ltd. Method and system for facilitating information searching on electronic devices
US7412708B1 (en) 2004-03-31 2008-08-12 Google Inc. Methods and systems for capturing information
US20080221987A1 (en) * 2007-03-07 2008-09-11 Ebay Inc. System and method for contextual advertisement and merchandizing based on an automatically generated user demographic profile
US20080225870A1 (en) * 2007-03-15 2008-09-18 Sundstrom Robert J Methods, systems, and computer program products for providing predicted likelihood of communication between users
US20080243631A1 (en) * 2007-03-30 2008-10-02 Kane Francis J Services for providing item association data
US20080243826A1 (en) * 2007-03-30 2008-10-02 Yahoo! Inc. System and method for determining semantically related terms
US20080243632A1 (en) * 2007-03-30 2008-10-02 Kane Francis J Service for providing item recommendations
US20080243480A1 (en) * 2007-03-30 2008-10-02 Yahoo! Inc. System and method for determining semantically related terms
US20080250323A1 (en) * 2007-04-04 2008-10-09 Huff Gerald B Method and apparatus for recommending an application-feature to a user
WO2008121737A1 (en) * 2007-03-30 2008-10-09 Amazon Technologies, Inc. Service for providing item recommendations
US20080270333A1 (en) * 2007-04-30 2008-10-30 Yahoo! Inc. System and method for determining semantically related terms using an active learning framework
US20080289039A1 (en) * 2007-05-18 2008-11-20 Sap Ag Method and system for protecting a message from an xml attack when being exchanged in a distributed and decentralized network system
US20080307053A1 (en) * 2007-06-08 2008-12-11 Mitnick Craig R System and Method for Permitting Geographically-Pertinent Information to be Ranked by Users According to Users' Geographic Proximity to Information and to Each Other for Affecting the Ranking of Such Information
US20080313180A1 (en) * 2007-06-14 2008-12-18 Microsoft Corporation Identification of topics for online discussions based on language patterns
US20090006398A1 (en) * 2007-06-29 2009-01-01 Shing Yan Lam Recommendation system with multiple integrated recommenders
US20090018904A1 (en) * 2007-07-09 2009-01-15 Ebay Inc. System and method for contextual advertising and merchandizing based on user configurable preferences
EP2020796A1 (en) * 2006-12-15 2009-02-04 Huawei Technologies Co Ltd A method for accessing user data and a user profile management server
US20090037399A1 (en) * 2007-07-31 2009-02-05 Yahoo! Inc. System and Method for Determining Semantically Related Terms
US20090037566A1 (en) * 2005-03-31 2009-02-05 British Telecommunications Public Limited Company Computer Network
EP2023634A1 (en) * 2007-08-10 2009-02-11 Kabushiki Kaisha Toshiba Program searching apparatus and program searching method
US20090063408A1 (en) * 2007-08-28 2009-03-05 International Business Machines Corporation Managing user ratings in a web services environment
US20090083388A1 (en) * 2003-12-10 2009-03-26 International Business Machines Corporation System for Information Capture
US20090100043A1 (en) * 2007-10-12 2009-04-16 Palo Alto Research Center Incorporated System And Method For Providing Orientation Into Digital Information
US20090099996A1 (en) * 2007-10-12 2009-04-16 Palo Alto Research Center Incorporated System And Method For Performing Discovery Of Digital Information In A Subject Area
US20090099839A1 (en) * 2007-10-12 2009-04-16 Palo Alto Research Center Incorporated System And Method For Prospecting Digital Information
US20090125934A1 (en) * 2007-11-11 2009-05-14 Microsoft Corporation User rating mechanism for media content
US7536408B2 (en) 2004-07-26 2009-05-19 Google Inc. Phrase-based indexing in an information retrieval system
US20090132526A1 (en) * 2007-11-19 2009-05-21 Jong-Hun Park Content recommendation apparatus and method using tag cloud
US20090144654A1 (en) * 2007-10-03 2009-06-04 Robert Brouwer Methods and apparatus for facilitating content consumption
US20090171754A1 (en) * 2007-12-28 2009-07-02 Kane Francis J Widget-assisted detection and exposure of cross-site behavioral associations
US20090171968A1 (en) * 2007-12-28 2009-07-02 Kane Francis J Widget-assisted content personalization based on user behaviors tracked across multiple web sites
US20090171755A1 (en) * 2007-12-28 2009-07-02 Kane Francis J Behavior-based generation of site-to-site referrals
US20090172551A1 (en) * 2007-12-28 2009-07-02 Kane Francis J Behavior-based selection of items to present on affiliate sites
US20090172021A1 (en) * 2007-12-28 2009-07-02 Kane Francis J Recommendations based on actions performed on multiple remote servers
US7581227B1 (en) 2004-03-31 2009-08-25 Google Inc. Systems and methods of synchronizing indexes
US7580929B2 (en) * 2004-07-26 2009-08-25 Google Inc. Phrase-based personalization of searches in an information retrieval system
US7580921B2 (en) 2004-07-26 2009-08-25 Google Inc. Phrase identification in an information retrieval system
US7584175B2 (en) 2004-07-26 2009-09-01 Google Inc. Phrase-based generation of document descriptions
US20090222735A1 (en) * 2008-02-28 2009-09-03 Clark Bryan W Systems and methods for enhancing browser history using syndicated data feeds
US20090240769A1 (en) * 2002-02-25 2009-09-24 Ascentive Llc Method and system for screening remote site connections and filtering data based on a community trust assessment
US7599914B2 (en) 2004-07-26 2009-10-06 Google Inc. Phrase-based searching in an information retrieval system
US20090254838A1 (en) * 2008-04-03 2009-10-08 Icurrent, Inc. Information display system based on user profile data with assisted and explicit profile modification
US20090292584A1 (en) * 2008-05-23 2009-11-26 Ebay Inc. System and method for context and community based customization for a user experience
US20090292743A1 (en) * 2008-05-21 2009-11-26 Bigus Joseph P Modeling user access to computer resources
US20100005424A1 (en) * 2008-03-05 2010-01-07 Neelakantan Sundaresan Virtual world system supporting a consumer experience
US20100030713A1 (en) * 2006-05-24 2010-02-04 Icom Limited Content engine
US20100030894A1 (en) * 2002-03-07 2010-02-04 David Cancel Computer program product and method for estimating internet traffic
US20100042460A1 (en) * 2008-08-12 2010-02-18 Kane Jr Francis J System for obtaining recommendations from multiple recommenders
US20100042608A1 (en) * 2008-08-12 2010-02-18 Kane Jr Francis J System for obtaining recommendations from multiple recommenders
US20100057536A1 (en) * 2008-08-28 2010-03-04 Palo Alto Research Center Incorporated System And Method For Providing Community-Based Advertising Term Disambiguation
US20100057577A1 (en) * 2008-08-28 2010-03-04 Palo Alto Research Center Incorporated System And Method For Providing Topic-Guided Broadening Of Advertising Targets In Social Indexing
US20100057716A1 (en) * 2008-08-28 2010-03-04 Stefik Mark J System And Method For Providing A Topic-Directed Search
US20100058195A1 (en) * 2008-08-28 2010-03-04 Palo Alto Research Center Incorporated System And Method For Interfacing A Web Browser Widget With Social Indexing
US7680809B2 (en) 2004-03-31 2010-03-16 Google Inc. Profile based capture component
US7680888B1 (en) 2004-03-31 2010-03-16 Google Inc. Methods and systems for processing instant messenger messages
US20100083217A1 (en) * 2008-09-30 2010-04-01 Dalal Vipul C System and method for orchestration of customization for a user expereince
US7693813B1 (en) 2007-03-30 2010-04-06 Google Inc. Index server architecture using tiered and sharded phrase posting lists
US20100094859A1 (en) * 2006-08-09 2010-04-15 Gupta Puneet K Folksonomy-Enhanced Enterprise-Centric Collaboration and Knowledge Management System
US7702614B1 (en) 2007-03-30 2010-04-20 Google Inc. Index updating using segment swapping
US7702618B1 (en) 2004-07-26 2010-04-20 Google Inc. Information retrieval system for archiving multiple document versions
EP2176730A1 (en) * 2007-08-08 2010-04-21 Baynote, Inc. Method and apparatus for context-based content recommendation
US20100106703A1 (en) * 2006-05-02 2010-04-29 Mark Cramer Dynamic search engine results employing user behavior
US7711679B2 (en) 2004-07-26 2010-05-04 Google Inc. Phrase-based detection of duplicate documents in an information retrieval system
US20100114654A1 (en) * 2008-10-31 2010-05-06 Hewlett-Packard Development Company, L.P. Learning user purchase intent from user-centric data
US20100122257A1 (en) * 2007-01-26 2010-05-13 Kyocera Corporation Electronic device and electronic device control method
US20100125540A1 (en) * 2008-11-14 2010-05-20 Palo Alto Research Center Incorporated System And Method For Providing Robust Topic Identification In Social Indexes
US20100125597A1 (en) * 2008-11-14 2010-05-20 Yahoo! Inc. System and method for determining search terms for use in sponsored searches
US7725508B2 (en) 2004-03-31 2010-05-25 Google Inc. Methods and systems for information capture and retrieval
US20100131333A1 (en) * 2007-08-06 2010-05-27 Keith Ball Method and system for product services analysis and optimization
US20100145902A1 (en) * 2008-12-09 2010-06-10 Ita Software, Inc. Methods and systems to train models to extract and integrate information from data sources
US20100180200A1 (en) * 2009-01-12 2010-07-15 Thierry Donneau-Golencer Electronic assistant
US20100185473A1 (en) * 2009-01-20 2010-07-22 Microsoft Corporation Document vault and application platform
US20100191742A1 (en) * 2009-01-27 2010-07-29 Palo Alto Research Center Incorporated System And Method For Managing User Attention By Detecting Hot And Cold Topics In Social Indexes
US20100191773A1 (en) * 2009-01-27 2010-07-29 Palo Alto Research Center Incorporated System And Method For Providing Default Hierarchical Training For Social Indexing
US20100191741A1 (en) * 2009-01-27 2010-07-29 Palo Alto Research Center Incorporated System And Method For Using Banded Topic Relevance And Time For Article Prioritization
US20100211568A1 (en) * 2009-02-19 2010-08-19 Wei Chu Personalized recommendations on dynamic content
US20100217722A1 (en) * 2006-08-28 2010-08-26 Manyworlds, Inc. Mutual Interest Inferencing System and Method
US7827125B1 (en) * 2006-06-01 2010-11-02 Trovix, Inc. Learning based on feedback for contextual personalized information retrieval
US20100280906A1 (en) * 2000-07-18 2010-11-04 Lim Stanley T Economic Filtering System for Delivery of Permission Based, Targeted, Incentivized Advertising
US7840691B1 (en) 2000-09-07 2010-11-23 Zamora Radio, Llc Personal broadcast server system for providing a customized broadcast
US20100296728A1 (en) * 2009-05-22 2010-11-25 Shinya Ohtani Discrimination Apparatus, Method of Discrimination, and Computer Program
US20100306079A1 (en) * 2003-03-31 2010-12-02 Sap Ag Collaborative Product Taxonomy Instantiation
US7853606B1 (en) 2004-09-14 2010-12-14 Google, Inc. Alternate methods of displaying search results
EP2271057A1 (en) * 2009-06-26 2011-01-05 Alcatel Lucent Device and method for exploiting exchanged data and actions of user communication equipment(s) to improve browsing dynamically, and associated network equipment
USRE42101E1 (en) * 2000-04-14 2011-02-01 Realnetworks, Inc. System and method of managing metadata data
US20110035381A1 (en) * 2008-04-23 2011-02-10 Simon Giles Thompson Method
US7890451B2 (en) 2002-10-09 2011-02-15 Compete, Inc. Computer program product and method for refining an estimate of internet traffic
US7925655B1 (en) 2007-03-30 2011-04-12 Google Inc. Query scheduling using hierarchical tiers of index servers
US20110093341A1 (en) * 2005-05-16 2011-04-21 Manyworlds, Inc. Explanatory Advertising Systems and Methods
US20110112986A1 (en) * 2005-01-18 2011-05-12 Manyworlds, Inc. Generative Investment Method and System
EP2328344A1 (en) * 2008-09-23 2011-06-01 Huawei Device Co., Ltd. Method, apparatus and system for playing programs
US7984056B1 (en) 2007-12-28 2011-07-19 Amazon Technologies, Inc. System for facilitating discovery and management of feeds
US8019744B1 (en) * 2004-10-06 2011-09-13 Shopzilla, Inc. Search ranking estimation
US8024323B1 (en) * 2003-11-13 2011-09-20 AudienceScience Inc. Natural language search for audience
US20110246174A1 (en) * 2008-01-17 2011-10-06 Geacom, Inc. Method and system for situational language interpretation
US8060816B1 (en) * 2000-10-31 2011-11-15 International Business Machines Corporation Methods and apparatus for intelligent crawling on the world wide web
US20110282861A1 (en) * 2010-05-11 2011-11-17 Microsoft Corporation Extracting higher-order knowledge from structured data
US8065254B1 (en) 2007-02-19 2011-11-22 Google Inc. Presenting a diversity of recommendations
US20110295612A1 (en) * 2010-05-28 2011-12-01 Thierry Donneau-Golencer Method and apparatus for user modelization
US8086594B1 (en) 2007-03-30 2011-12-27 Google Inc. Bifurcated document relevance scoring
US8087051B2 (en) 2000-06-30 2011-12-27 Thomson Licensing Database management system and method for electronic program guide and television channel lineup organization
US8090794B1 (en) * 2008-08-25 2012-01-03 Intuit Inc. Technique for customizing displayed content
US8099407B2 (en) 2004-03-31 2012-01-17 Google Inc. Methods and systems for processing media files
US8103659B1 (en) * 2005-06-06 2012-01-24 A9.Com, Inc. Perspective-based item navigation
US8115869B2 (en) 2007-02-28 2012-02-14 Samsung Electronics Co., Ltd. Method and system for extracting relevant information from content metadata
US8117193B2 (en) 2007-12-21 2012-02-14 Lemi Technology, Llc Tunersphere
US8117223B2 (en) 2007-09-07 2012-02-14 Google Inc. Integrating external related phrase information into a phrase-based indexing information retrieval system
US8122020B1 (en) 2005-11-17 2012-02-21 Amazon Technologies, Inc. Recommendations based on item tagging activities of users
US8126865B1 (en) * 2003-12-31 2012-02-28 Google Inc. Systems and methods for syndicating and hosting customized news content
US8135718B1 (en) 2007-02-16 2012-03-13 Google Inc. Collaborative filtering
US8140524B1 (en) 2005-03-30 2012-03-20 Google Inc. Estimating confidence for query revision models
US8156444B1 (en) 2003-12-31 2012-04-10 Google Inc. Systems and methods for determining a user interface attribute
US8161053B1 (en) 2004-03-31 2012-04-17 Google Inc. Methods and systems for eliminating duplicate events
US20120095861A1 (en) * 2010-10-15 2012-04-19 At&T Intellectual Property I, L.P. Personal customer care agent
US8166045B1 (en) 2007-03-30 2012-04-24 Google Inc. Phrase extraction using subphrase scoring
US8166021B1 (en) 2007-03-30 2012-04-24 Google Inc. Query phrasification
US8176068B2 (en) 2007-10-31 2012-05-08 Samsung Electronics Co., Ltd. Method and system for suggesting search queries on electronic devices
US8190989B1 (en) 2003-04-29 2012-05-29 Google Inc. Methods and apparatus for assisting in completion of a form
US20120136886A1 (en) * 2003-12-30 2012-05-31 Microsoft Corporation Incremental query refinement
US8195799B1 (en) * 2011-10-26 2012-06-05 SHTC Holdings LLC Smart test article optimizer
US8200688B2 (en) * 2006-03-07 2012-06-12 Samsung Electronics Co., Ltd. Method and system for facilitating information searching on electronic devices
US8209724B2 (en) 2007-04-25 2012-06-26 Samsung Electronics Co., Ltd. Method and system for providing access to information of potential interest to a user
US20120191776A1 (en) * 2011-01-20 2012-07-26 Linkedin Corporation Methods and systems for recommending a context based on content interaction
US20120226750A1 (en) * 2011-03-01 2012-09-06 Rex Gibson Online peer review of internet data
US8275839B2 (en) 2004-03-31 2012-09-25 Google Inc. Methods and systems for processing email messages
EP2506522A1 (en) * 2010-08-04 2012-10-03 Tencent Technology (Shenzhen) Company Limited Method and device for pushing data
US8302127B2 (en) 2000-09-25 2012-10-30 Thomson Licensing System and method for personalized TV
US8316015B2 (en) 2007-12-21 2012-11-20 Lemi Technology, Llc Tunersphere
US20120323876A1 (en) * 2011-06-16 2012-12-20 Microsoft Corporation Search results based on user and result profiles
US8346777B1 (en) 2004-03-31 2013-01-01 Google Inc. Systems and methods for selectively storing event data
CN102855248A (en) * 2011-06-29 2013-01-02 中国移动通信集团广西有限公司 Determination method, apparatus and system for user characteristic information
WO2013001147A1 (en) * 2011-06-27 2013-01-03 Nokia Corporation Method and apparatus for providing recommendations based on locally generated models
US8351763B2 (en) 1998-06-12 2013-01-08 Thomson Licensing Television program recording with user preference determination
US8352984B2 (en) 1998-06-12 2013-01-08 Thomson Licensing System and method for generating and managing user preference information for scheduled and stored television programs
US8386728B1 (en) 2004-03-31 2013-02-26 Google Inc. Methods and systems for prioritizing a crawl
US20130054410A1 (en) * 2001-07-17 2013-02-28 Incucomm, Incorporated System and Method for Providing Requested Information to Thin Clients
US8402387B1 (en) 2008-04-24 2013-03-19 Google Inc. Advertising based on user models
US8407226B1 (en) 2007-02-16 2013-03-26 Google Inc. Collaborative filtering
US20130086071A1 (en) * 2011-09-30 2013-04-04 Jive Software, Inc. Augmenting search with association information
US20130132851A1 (en) * 2011-11-22 2013-05-23 International Business Machines Corporation Sentiment estimation of web browsing user
US8463930B2 (en) 2009-04-29 2013-06-11 Lemi Technology, Llc Skip feature for a broadcast or multicast media station
US20130182835A1 (en) * 1999-07-09 2013-07-18 Jpmorgan Chase Bank, N.A. System and Method of Intelligent Call Routing for Cross Sell Offer Selection based on Optimization Parameters or Account-level Data
US8494899B2 (en) 2008-12-02 2013-07-23 Lemi Technology, Llc Dynamic talk radio program scheduling
US8510453B2 (en) 2007-03-21 2013-08-13 Samsung Electronics Co., Ltd. Framework for correlating content on a local network with information on an external network
US8515998B1 (en) 2001-03-07 2013-08-20 Bascom Research LLP Framework for managing document objects stored on a network
US8521731B2 (en) 2008-07-09 2013-08-27 Yahoo! Inc. Systems and methods for query expansion in sponsored search
US20130238783A1 (en) * 2012-03-09 2013-09-12 Oracle International Corporation Personalized navigation shortcuts
US8538892B2 (en) 2001-06-21 2013-09-17 Thomson Licenging Method and system for electronic purchases using an intelligent data carrier medium, electronic coupon system, and interactive TV infrastructure
US8543570B1 (en) * 2008-06-10 2013-09-24 Surf Canyon Incorporated Adaptive user interface for real-time search relevance feedback
US20130268317A1 (en) * 2010-12-07 2013-10-10 Digital Foodie Oy Arrangement for facilitating shopping and related method
US8566263B2 (en) 2003-11-28 2013-10-22 World Assets Consulting Ag, Llc Adaptive computer-based personalities
USRE44559E1 (en) 2003-11-28 2013-10-22 World Assets Consulting Ag, Llc Adaptive social computing methods
US20130282361A1 (en) * 2012-04-20 2013-10-24 Sap Ag Obtaining data from electronic documents
US8595214B1 (en) 2004-03-31 2013-11-26 Google Inc. Systems and methods for article location and retrieval
US8595228B1 (en) 2009-01-09 2013-11-26 Google Inc. Preferred sites
US8600920B2 (en) 2003-11-28 2013-12-03 World Assets Consulting Ag, Llc Affinity propagation in adaptive network-based systems
US8600926B2 (en) 2011-03-29 2013-12-03 Manyworlds, Inc. Integrated interest and expertise-based discovery system and method
US8631076B1 (en) 2004-03-31 2014-01-14 Google Inc. Methods and systems for associating instant messenger events
US8635216B1 (en) * 2004-09-30 2014-01-21 Avaya Inc. Enhancing network information retrieval according to a user search profile
US8645292B2 (en) 2011-03-29 2014-02-04 Manyworlds, Inc. Serendipitous recommendations system and method
US20140067369A1 (en) * 2012-08-30 2014-03-06 Xerox Corporation Methods and systems for acquiring user related information using natural language processing techniques
US8676837B2 (en) 2003-12-31 2014-03-18 Google Inc. Systems and methods for personalizing aggregated news content
US20140143012A1 (en) * 2012-11-21 2014-05-22 Insightera Ltd. Method and system for predictive marketing campigns based on users online behavior and profile
CN103842997A (en) * 2011-08-04 2014-06-04 K·波波夫 Searching for and creating an adaptive content
US8751507B2 (en) * 2007-06-29 2014-06-10 Amazon Technologies, Inc. Recommendation system with multiple integrated recommenders
US8755763B2 (en) 1998-01-22 2014-06-17 Black Hills Media Method and device for an internet radio capable of obtaining playlist content from a content server
USRE44966E1 (en) 2003-11-28 2014-06-24 World Assets Consulting Ag, Llc Adaptive recommendations systems
USRE44968E1 (en) 2003-11-28 2014-06-24 World Assets Consulting Ag, Llc Adaptive self-modifying and recombinant systems
USRE44967E1 (en) 2003-11-28 2014-06-24 World Assets Consulting Ag, Llc Adaptive social and process network systems
US20140189525A1 (en) * 2012-12-28 2014-07-03 Yahoo! Inc. User behavior models based on source domain
US20140222784A1 (en) * 2012-12-22 2014-08-07 Mmodal Ip Llc User Interface for Predictive Model Generation
US8806047B2 (en) 2009-04-29 2014-08-12 Lemi Technology, Llc Skip feature for a broadcast or multicast media station
US20140236953A1 (en) * 2009-02-11 2014-08-21 Jeffrey A. Rapaport Methods using social topical adaptive networking system
US8843467B2 (en) 2007-05-15 2014-09-23 Samsung Electronics Co., Ltd. Method and system for providing relevant information to a user of a device in a local network
US8843433B2 (en) 2011-03-29 2014-09-23 Manyworlds, Inc. Integrated search and adaptive discovery system and method
US8856670B1 (en) * 2009-08-25 2014-10-07 Intuit Inc. Technique for customizing a user interface
US8863221B2 (en) 2006-03-07 2014-10-14 Samsung Electronics Co., Ltd. Method and system for integrating content and services among multiple networks
US8868590B1 (en) 2011-11-17 2014-10-21 Sri International Method and system utilizing a personalized user model to develop a search request
US8874567B2 (en) 2004-03-29 2014-10-28 Google Inc. Variable personalization of search results in a search engine
US8935269B2 (en) 2006-12-04 2015-01-13 Samsung Electronics Co., Ltd. Method and apparatus for contextual search and query refinement on consumer electronics devices
US8938465B2 (en) 2008-09-10 2015-01-20 Samsung Electronics Co., Ltd. Method and system for utilizing packaged content sources to identify and provide information based on contextual information
US8954420B1 (en) 2003-12-31 2015-02-10 Google Inc. Methods and systems for improving a search ranking using article information
US8954580B2 (en) 2012-01-27 2015-02-10 Compete, Inc. Hybrid internet traffic measurement using site-centric and panel data
US8977573B2 (en) 2012-03-01 2015-03-10 Nice-Systems Ltd. System and method for identifying customers in social media
US8976955B2 (en) 2011-11-28 2015-03-10 Nice-Systems Ltd. System and method for tracking web interactions with real time analytics
US9003015B2 (en) 2010-12-21 2015-04-07 Sitecore A/S Method and a system for managing a website using profile key patterns
US9003441B1 (en) * 2005-04-13 2015-04-07 Google Inc. Probabilistic inference of demographic information from user selection of content
US9015159B1 (en) * 2002-03-29 2015-04-21 Google Inc. Method for searching media
US9015147B2 (en) 2007-12-20 2015-04-21 Porto Technology, Llc System and method for generating dynamically filtered content results, including for audio and/or video channels
US20150112918A1 (en) * 2012-03-17 2015-04-23 Beijing Yidian Wangju Technology Co., Ltd. Method and system for recommending content to a user
US9031944B2 (en) 2010-04-30 2015-05-12 Palo Alto Research Center Incorporated System and method for providing multi-core and multi-level topical organization in social indexes
US9043352B1 (en) 2001-03-07 2015-05-26 Bascom Research, Llc Method for searching document objects on a network
US20150161292A1 (en) * 2013-12-05 2015-06-11 Richplay Information Co., Ltd. Method for recommending document
US9075861B2 (en) 2006-03-06 2015-07-07 Veveo, Inc. Methods and systems for segmenting relative user preferences into fine-grain and coarse-grain collections
US20150193789A1 (en) * 2014-01-03 2015-07-09 Mastercard International Incorporated Method and system for personalized news recommendations based on purchase behavior
US9105028B2 (en) 2005-08-10 2015-08-11 Compete, Inc. Monitoring clickstream behavior of viewers of online advertisements and search results
US9141589B2 (en) * 2005-03-09 2015-09-22 Google Inc. Method and an apparatus to provide a personalized page
US9160680B1 (en) 2014-11-18 2015-10-13 Kaspersky Lab Zao System and method for dynamic network resource categorization re-assignment
USRE45770E1 (en) 2003-11-28 2015-10-20 World Assets Consulting Ag, Llc Adaptive recommendation explanations
US9195640B1 (en) 2009-01-12 2015-11-24 Sri International Method and system for finding content having a desired similarity
US9203912B2 (en) 2007-11-14 2015-12-01 Qualcomm Incorporated Method and system for message value calculation in a mobile environment
US9223868B2 (en) * 2004-06-28 2015-12-29 Google Inc. Deriving and using interaction profiles
US9229900B2 (en) 2011-01-20 2016-01-05 Linkedin Corporation Techniques for ascribing social attributes to content
US9245033B2 (en) 2009-04-02 2016-01-26 Graham Holdings Company Channel sharing
US9262446B1 (en) 2005-12-29 2016-02-16 Google Inc. Dynamically ranking entries in a personal data book
US9262681B1 (en) * 2014-11-19 2016-02-16 Amazon Technologies, Inc. Probabilistic registration of interactions, actions or activities from multiple views
WO2016025658A1 (en) * 2014-08-15 2016-02-18 Microsoft Technology Licensing, Llc Auto recognition of acquirable entities
US9286385B2 (en) 2007-04-25 2016-03-15 Samsung Electronics Co., Ltd. Method and system for providing access to information of potential interest to a user
USRE45952E1 (en) 2002-03-29 2016-03-29 Google Inc. Method for searching media
US9392074B2 (en) 2007-07-07 2016-07-12 Qualcomm Incorporated User profile generation architecture for mobile content-message targeting
US9391789B2 (en) 2007-12-14 2016-07-12 Qualcomm Incorporated Method and system for multi-level distribution information cache management in a mobile environment
US9398113B2 (en) 2007-07-07 2016-07-19 Qualcomm Incorporated Methods and systems for providing targeted information using identity masking in a wireless communications device
US9430783B1 (en) 2014-06-13 2016-08-30 Snapchat, Inc. Prioritization of messages within gallery
US9443022B2 (en) 2006-06-05 2016-09-13 Google Inc. Method, system, and graphical user interface for providing personalized recommendations of popular search queries
US9483568B1 (en) 2013-06-05 2016-11-01 Google Inc. Indexing system
US9497498B2 (en) * 2015-01-23 2016-11-15 Robert Hain System and method for live streaming of content
US9501506B1 (en) 2013-03-15 2016-11-22 Google Inc. Indexing system
US9516370B1 (en) 2004-05-05 2016-12-06 Black Hills Media, Llc Method, device, and system for directing a wireless speaker from a mobile phone to receive and render a playlist from a content server on the internet
US9552555B1 (en) * 2012-12-31 2017-01-24 Google Inc. Methods, systems, and media for recommending content items based on topics
WO2017035970A1 (en) * 2015-08-31 2017-03-09 北京百度网讯科技有限公司 Information pushing method and apparatus
US9628573B1 (en) 2012-05-01 2017-04-18 Amazon Technologies, Inc. Location-based interaction with digital works
US9665630B1 (en) * 2012-06-18 2017-05-30 EMC IP Holding Company LLC Techniques for providing storage hints for use in connection with data movement optimizations
US9672475B2 (en) 2014-12-19 2017-06-06 International Business Machines Corporation Automated opinion prediction based on indirect information
US9715496B1 (en) * 2016-07-08 2017-07-25 Asapp, Inc. Automatically responding to a request of a user
US9734468B2 (en) 2012-02-21 2017-08-15 Nice Ltd. System and method for resolving customer communications
US9736524B2 (en) 2011-01-06 2017-08-15 Veveo, Inc. Methods of and systems for content search based on environment sampling
US9749693B2 (en) 2006-03-24 2017-08-29 Rovi Guides, Inc. Interactive media guidance application with intelligent navigation and display features
US9754288B2 (en) 2009-06-30 2017-09-05 Amazon Technologies, Inc. Recommendation of media content items based on geolocation and venue
US9769208B2 (en) 2015-05-28 2017-09-19 International Business Machines Corporation Inferring security policies from semantic attributes
US9785987B2 (en) 2010-04-22 2017-10-10 Microsoft Technology Licensing, Llc User interface for information presentation system
US9785888B2 (en) * 2010-05-31 2017-10-10 Sony Corporation Information processing apparatus, information processing method, and program for prediction model generated based on evaluation information
US9805127B2 (en) 2011-01-20 2017-10-31 Linkedin Corporation Methods and systems for utilizing activity data with clustered events
US9866999B1 (en) 2014-01-12 2018-01-09 Investment Asset Holdings Llc Location-based messaging
US9900395B2 (en) 2012-01-27 2018-02-20 Comscore, Inc. Dynamic normalization of internet traffic
EP3163465A4 (en) * 2014-06-25 2018-03-14 Beijing Baidupay Science And Tech Co Ltd Method and device for data mining based on user's search behaviour

Families Citing this family (189)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7082426B2 (en) * 1993-06-18 2006-07-25 Cnet Networks, Inc. Content aggregation method and apparatus for an on-line product catalog
US7233977B2 (en) * 1998-12-18 2007-06-19 Emc Corporation Messaging mechanism employing mailboxes for inter processor communications
DE10121791B4 (en) * 2000-05-18 2004-11-25 International Business Machines Corporation Method and apparatus for dynamic web page order
US8255479B2 (en) * 2000-08-03 2012-08-28 Qtrax, Inc Online network and associated methods
US8255791B2 (en) * 2000-11-29 2012-08-28 Dov Koren Collaborative, flexible, interactive real-time displays
US7197708B1 (en) * 2000-12-20 2007-03-27 Cisco Technology, Inc. Set top to internet device communication and use
JP3997774B2 (en) * 2001-12-11 2007-10-24 ソニー株式会社 Data processing system, data processing method, and an information processing apparatus, and computer program
US20030131070A1 (en) * 2002-01-10 2003-07-10 International Business Machines Corporation Customization of information from web sites
US7324987B2 (en) * 2002-10-23 2008-01-29 Infonow Corporation System and method for improving resolution of channel data
US7630916B2 (en) * 2003-06-25 2009-12-08 Microsoft Corporation Systems and methods for improving collaborative filtering
EP1668459A4 (en) * 2003-09-22 2009-05-13 Citicorp Credit Services Inc Method and system for purchase-based segmentation
US7526459B2 (en) * 2004-11-04 2009-04-28 Manyworlds, Inc. Adaptive social and process network systems
EP2662784A1 (en) * 2004-03-15 2013-11-13 Yahoo! Inc. Search systems and methods with integration of user annotations
US8666746B2 (en) * 2004-05-13 2014-03-04 At&T Intellectual Property Ii, L.P. System and method for generating customized text-to-speech voices
US7302406B2 (en) * 2004-06-17 2007-11-27 Internation Business Machines Corporation Method, apparatus and system for retrieval of specialized consumer information
JP4653157B2 (en) * 2005-01-25 2011-03-16 株式会社ターボデータラボラトリー Tree search, aggregation, sorting method, information processing apparatus, and a search of the tree, aggregate, sort program
US20060190225A1 (en) * 2005-02-18 2006-08-24 Brand Matthew E Collaborative filtering using random walks of Markov chains
JP4829563B2 (en) * 2005-08-03 2011-12-07 キヤノン株式会社 Control method and control apparatus
US7711737B2 (en) * 2005-09-12 2010-05-04 Microsoft Corporation Multi-document keyphrase extraction using partial mutual information
US20070112852A1 (en) * 2005-11-07 2007-05-17 Nokia Corporation Methods for characterizing content item groups
US8364467B1 (en) 2006-03-31 2013-01-29 Google Inc. Content-based classification
US20070260626A1 (en) * 2006-05-04 2007-11-08 Claudia Reisz Method for customer-choice-based bundling of product options
US20070266025A1 (en) * 2006-05-12 2007-11-15 Microsoft Corporation Implicit tokenized result ranking
US20070265999A1 (en) * 2006-05-15 2007-11-15 Einat Amitay Search Performance and User Interaction Monitoring of Search Engines
US7890533B2 (en) * 2006-05-17 2011-02-15 Noblis, Inc. Method and system for information extraction and modeling
US9547648B2 (en) * 2006-08-03 2017-01-17 Excalibur Ip, Llc Electronic document information extraction
US7895209B2 (en) * 2006-09-11 2011-02-22 Microsoft Corporation Presentation of information based on current activity
JP5233220B2 (en) * 2006-10-11 2013-07-10 株式会社リコー Page additional information sharing management method
US20080228675A1 (en) * 2006-10-13 2008-09-18 Move, Inc. Multi-tiered cascading crawling system
US20080091423A1 (en) * 2006-10-13 2008-04-17 Shourya Roy Generation of domain models from noisy transcriptions
CN101589360A (en) * 2006-10-18 2009-11-25 谷歌公司 Generic online ranking system and method suitable for syndication
WO2008055034A3 (en) 2006-10-30 2008-09-04 Noblis Inc Method and system for personal information extraction and modeling with fully generalized extraction contexts
US9519715B2 (en) * 2006-11-02 2016-12-13 Excalibur Ip, Llc Personalized search
CN101216826B (en) * 2007-01-05 2011-06-08 鸿富锦精密工业(深圳)有限公司 Information search system and method
US7844602B2 (en) * 2007-01-19 2010-11-30 Healthline Networks, Inc. Method and system for establishing document relevance
US8200663B2 (en) 2007-04-25 2012-06-12 Chacha Search, Inc. Method and system for improvement of relevance of search results
US8108409B2 (en) * 2007-07-19 2012-01-31 Hewlett-Packard Development Company, L.P. Determining top combinations of items to present to a user
US7979321B2 (en) * 2007-07-25 2011-07-12 Ebay Inc. Merchandising items of topical interest
US20090063168A1 (en) * 2007-08-29 2009-03-05 Finn Peter G Conducting marketing activity in relation to a virtual world based on monitored virtual world activity
US8909635B2 (en) 2007-09-21 2014-12-09 International Business Machines Corporation Profiling content creation and retrieval in a content management system
US8126826B2 (en) 2007-09-21 2012-02-28 Noblis, Inc. Method and system for active learning screening process with dynamic information modeling
US20090100032A1 (en) * 2007-10-12 2009-04-16 Chacha Search, Inc. Method and system for creation of user/guide profile in a human-aided search system
US8112407B2 (en) * 2007-10-24 2012-02-07 The Invention Science Fund I, Llc Selecting a second content based on a user's reaction to a first content
US20090112695A1 (en) * 2007-10-24 2009-04-30 Searete Llc, A Limited Liability Corporation Of The State Of Delaware Physiological response based targeted advertising
US8234262B2 (en) * 2007-10-24 2012-07-31 The Invention Science Fund I, Llc Method of selecting a second content based on a user's reaction to a first content of at least two instances of displayed content
US20090112849A1 (en) * 2007-10-24 2009-04-30 Searete Llc Selecting a second content based on a user's reaction to a first content of at least two instances of displayed content
US8001108B2 (en) * 2007-10-24 2011-08-16 The Invention Science Fund I, Llc Returning a new content based on a person's reaction to at least two instances of previously displayed content
US8126867B2 (en) * 2007-10-24 2012-02-28 The Invention Science Fund I, Llc Returning a second content based on a user's reaction to a first content
US20090113297A1 (en) * 2007-10-24 2009-04-30 Searete Llc, A Limited Liability Corporation Of The State Of Delaware Requesting a second content based on a user's reaction to a first content
US20090112697A1 (en) * 2007-10-30 2009-04-30 Searete Llc, A Limited Liability Corporation Of The State Of Delaware Providing personalized advertising
US9513699B2 (en) * 2007-10-24 2016-12-06 Invention Science Fund I, LL Method of selecting a second content based on a user's reaction to a first content
US20090112696A1 (en) * 2007-10-24 2009-04-30 Jung Edward K Y Method of space-available advertising in a mobile device
US20090112694A1 (en) * 2007-10-24 2009-04-30 Searete Llc, A Limited Liability Corporation Of The State Of Delaware Targeted-advertising based on a sensed physiological response by a person to a general advertisement
US9582805B2 (en) * 2007-10-24 2017-02-28 Invention Science Fund I, Llc Returning a personalized advertisement
US20090112693A1 (en) * 2007-10-24 2009-04-30 Jung Edward K Y Providing personalized advertising
US7844555B2 (en) * 2007-11-13 2010-11-30 Microsoft Corporation Ranker selection for statistical natural language processing
US7925602B2 (en) * 2007-12-07 2011-04-12 Microsoft Corporation Maximum entropy model classfier that uses gaussian mean values
US8271357B2 (en) 2007-12-11 2012-09-18 Ebay Inc. Presenting items based on activity rates
US8032471B2 (en) * 2007-12-21 2011-10-04 Yahoo! Inc. System and method for annotating and analyzing reviews with inferred analytics
US9117219B2 (en) * 2007-12-31 2015-08-25 Peer 39 Inc. Method and a system for selecting advertising spots
US20100070501A1 (en) * 2008-01-15 2010-03-18 Walsh Paul J Enhancing and storing data for recall and use using user feedback
US20090182622A1 (en) * 2008-01-15 2009-07-16 Agarwal Amit D Enhancing and storing data for recall and use
US20150161175A1 (en) * 2008-02-08 2015-06-11 Google Inc. Alternative image queries
US9009600B2 (en) * 2008-02-29 2015-04-14 Red Hat, Inc. Filter module for a browser history
US8762364B2 (en) * 2008-03-18 2014-06-24 Yahoo! Inc. Personalizing sponsored search advertising layout using user behavior history
US8224712B2 (en) 2008-04-14 2012-07-17 Dell Products L.P. Point of sale personalization of an information handling system
US8184116B2 (en) * 2008-04-24 2012-05-22 International Business Machines Corporation Object based avatar tracking
US8466931B2 (en) * 2008-04-24 2013-06-18 International Business Machines Corporation Color modification of objects in a virtual universe
US8233005B2 (en) * 2008-04-24 2012-07-31 International Business Machines Corporation Object size modifications based on avatar distance
US8259100B2 (en) * 2008-04-24 2012-09-04 International Business Machines Corporation Fixed path transitions
US8212809B2 (en) * 2008-04-24 2012-07-03 International Business Machines Corporation Floating transitions
US20090313550A1 (en) * 2008-06-17 2009-12-17 Microsoft Corporation Theme Based Content Interaction
US20090327859A1 (en) * 2008-06-26 2009-12-31 Yahoo! Inc. Method and system for utilizing web document layout and presentation to improve user experience in web search
US8990705B2 (en) * 2008-07-01 2015-03-24 International Business Machines Corporation Color modifications of objects in a virtual universe based on user display settings
US8471843B2 (en) * 2008-07-07 2013-06-25 International Business Machines Corporation Geometric and texture modifications of objects in a virtual universe based on real world user characteristics
US8346532B2 (en) * 2008-07-11 2013-01-01 International Business Machines Corporation Managing the creation, detection, and maintenance of sensitive information
US8832098B2 (en) * 2008-07-29 2014-09-09 Yahoo! Inc. Research tool access based on research session detection
CN101661367A (en) * 2008-08-25 2010-03-03 扬智科技股份有限公司 Generation method of menu shortcut mode and user interface device
CN101661475B (en) * 2008-08-26 2013-04-24 华为技术有限公司 Search method and system
US8606651B2 (en) * 2008-09-05 2013-12-10 Sony Corporation Generation of home network use recommendations based on collected metadata of prior connected items
US20100082527A1 (en) * 2008-09-24 2010-04-01 Yahoo! Inc. Dated metadata to support multiple versions of user profiles for targeting of personalized content
US8521757B1 (en) 2008-09-26 2013-08-27 Symantec Corporation Method and apparatus for template-based processing of electronic documents
CA2738421A1 (en) * 2008-09-29 2010-04-01 Telefonaktiebolaget Lm Ericsson (Publ) Double weighted correlation scheme
US20100082684A1 (en) * 2008-10-01 2010-04-01 Yahoo! Inc. Method and system for providing personalized web experience
US8271875B2 (en) * 2008-10-07 2012-09-18 Google Inc. Providing help information
US20100125502A1 (en) * 2008-11-18 2010-05-20 Peer 39 Inc. Method and system for identifying web documents for advertisements
US20100125577A1 (en) * 2008-11-19 2010-05-20 International Business Machines Corporation Extending Distribution Lists
US9477672B2 (en) 2009-12-02 2016-10-25 Gartner, Inc. Implicit profile for use with recommendation engine and/or question router
US8234274B2 (en) * 2008-12-18 2012-07-31 Nec Laboratories America, Inc. Systems and methods for characterizing linked documents using a latent topic model
US20100169177A1 (en) * 2008-12-29 2010-07-01 Virgil Griffith Method and system for assessing behavior of a webpage visitor
US20100177117A1 (en) * 2009-01-14 2010-07-15 International Business Machines Corporation Contextual templates for modifying objects in a virtual universe
US9477763B2 (en) 2009-03-02 2016-10-25 Excalibur IP, LC Personalized search results utilizing previously navigated web sites
US9391825B1 (en) * 2009-03-24 2016-07-12 Amazon Technologies, Inc. System and method for tracking service results
US8195611B2 (en) * 2009-03-31 2012-06-05 International Business Machines Corporation Using a sparse file as a clone of a file
US9412127B2 (en) 2009-04-08 2016-08-09 Ebay Inc. Methods and systems for assessing the quality of an item listing
US8290972B1 (en) * 2009-04-29 2012-10-16 Netapp, Inc. System and method for storing and accessing data using a plurality of probabilistic data structures
US20100287462A1 (en) * 2009-05-05 2010-11-11 Paul A. Lipari System and method for content selection for web page indexing
US9442621B2 (en) * 2009-05-05 2016-09-13 Suboti, Llc System, method and computer readable medium for determining user attention area from user interface events
US20100287152A1 (en) 2009-05-05 2010-11-11 Paul A. Lipari System, method and computer readable medium for web crawling
US20100287031A1 (en) * 2009-05-07 2010-11-11 Mckenna Charles Method, Apparatus, System, and Computer Program for Selecting Replacement User Devices
US8583511B2 (en) * 2009-05-19 2013-11-12 Bradley Marshall Hendrickson Systems and methods for storing customer purchasing and preference data and enabling a customer to pre-register orders and events
US20100301114A1 (en) * 2009-05-26 2010-12-02 Lo Faro Walter F Method and system for transaction based profiling of customers within a merchant network
CN101901235B (en) 2009-05-27 2013-03-27 国际商业机器公司 Method and system for document processing
US8374990B2 (en) * 2009-06-08 2013-02-12 Business Objects Software Limited Situational resource rating
EP2261819A1 (en) * 2009-06-11 2010-12-15 Alcatel Lucent Device for providing a user profile dedicated to specific needs of a requesting service, and associated network equipment
US8510308B1 (en) * 2009-06-16 2013-08-13 Google Inc. Extracting semantic classes and instances from text
KR20110031087A (en) * 2009-09-18 2011-03-24 인터내셔널 비지네스 머신즈 코포레이션 Link clouds and user/community-driven dynamic interlinking of resources
EP2312515A1 (en) * 2009-10-16 2011-04-20 Alcatel Lucent Device for determining potential future interests to be introduced into profile(s) of user(s) of communication equipment(s)
US8122005B1 (en) * 2009-10-22 2012-02-21 Google Inc. Training set construction for taxonomic classification
US8972436B2 (en) * 2009-10-28 2015-03-03 Yahoo! Inc. Translation model and method for matching reviews to objects
US9886681B2 (en) 2009-11-24 2018-02-06 International Business Machines Corporation Creating an aggregate report of a presence of a user on a network
US20110131536A1 (en) * 2009-12-02 2011-06-02 Xerox Corporation Generating and ranking information units including documents associated with document environments
US8255380B2 (en) * 2009-12-18 2012-08-28 International Business Machines Corporation System and method for ontology-based location of expertise
US20170316088A9 (en) * 2010-02-03 2017-11-02 Gartner, Inc. Methods and systems for modifying a user profile for a recommendation algorithm and making recommendations based on user interactions with items
US20110213679A1 (en) * 2010-02-26 2011-09-01 Ebay Inc. Multi-quantity fixed price referral systems and methods
US9336377B2 (en) 2010-04-21 2016-05-10 Lexmark International Technology Sarl Synchronized sign-on methods for non-programmatic integration systems
US9081632B2 (en) 2010-04-21 2015-07-14 Lexmark International Technology Sa Collaboration methods for non-programmatic integration systems
US9183202B2 (en) * 2010-04-21 2015-11-10 Lexmark International Technology, SA Notification methods for non-programmatic integration systems
US20120089914A1 (en) * 2010-04-27 2012-04-12 Surfwax Inc. User interfaces for navigating structured content
US8584034B2 (en) 2010-04-27 2013-11-12 Surfwax, Inc. User interfaces for navigating structured content
CN101877720A (en) * 2010-05-25 2010-11-03 中电投远达环保工程有限公司;中冶赛迪工程技术股份有限公司 B/S structure-based discrete data point curve structural system and structural method thereof
US8583674B2 (en) * 2010-06-18 2013-11-12 Microsoft Corporation Media item recommendation
US9449282B2 (en) * 2010-07-01 2016-09-20 Match.Com, L.L.C. System for determining and optimizing for relevance in match-making systems
CN102411591A (en) 2010-09-21 2012-04-11 阿里巴巴集团控股有限公司 Method and equipment for processing information
US9535884B1 (en) 2010-09-30 2017-01-03 Amazon Technologies, Inc. Finding an end-of-body within content
US20120109947A1 (en) * 2010-11-02 2012-05-03 BrightEdge Technologies Multi-tenant analytics processing
CN102542474B (en) 2010-12-07 2015-10-21 阿里巴巴集团控股有限公司 Method and apparatus for sorting query results
US20120151339A1 (en) * 2010-12-10 2012-06-14 Microsoft Corporation Accessing and interacting with information
WO2012079254A1 (en) * 2010-12-17 2012-06-21 北京交通大学 Program recommending device and program recommending method
US9158775B1 (en) * 2010-12-18 2015-10-13 Google Inc. Scoring stream items in real time
US20120172025A1 (en) * 2010-12-29 2012-07-05 Nokia Corporation Method and apparatus for providing recommendations based on preloaded models
US8612419B2 (en) 2011-01-31 2013-12-17 International Business Machines Corporation Intelligent content discovery for content consumers
US20120203592A1 (en) * 2011-02-08 2012-08-09 Balaji Ravindran Methods, apparatus, and articles of manufacture to determine search engine market share
US9015141B2 (en) 2011-02-08 2015-04-21 The Nielsen Company (Us), Llc Methods, apparatus, and articles of manufacture to measure search results
US20120203733A1 (en) * 2011-02-09 2012-08-09 Zhang Amy H Method and system for personal cloud engine
US8996495B2 (en) * 2011-02-15 2015-03-31 Ebay Inc. Method and system for ranking search results based on category demand normalized using impressions
US9679561B2 (en) * 2011-03-28 2017-06-13 Nuance Communications, Inc. System and method for rapid customization of speech recognition models
US20120278268A1 (en) * 2011-04-28 2012-11-01 Nokia Corporation Method and apparatus for extracting explicit profile representation through adaptive recommender system
US9588632B1 (en) * 2011-05-24 2017-03-07 A9.Com, Inc. Techniques for providing personalized behavior-based content
US8577876B2 (en) * 2011-06-06 2013-11-05 Met Element, Inc. System and method for determining art preferences of people
US8938391B2 (en) 2011-06-12 2015-01-20 Microsoft Corporation Dynamically adding personalization features to language models for voice search
US9116979B2 (en) * 2011-06-22 2015-08-25 Rogers Communications Inc. Systems and methods for creating an interest profile for a user
CN102956009B (en) * 2011-08-16 2017-03-01 阿里巴巴集团控股有限公司 E-commerce information recommendation method and device based on user behavior
US8914496B1 (en) 2011-09-12 2014-12-16 Amazon Technologies, Inc. Tracking user behavior relative to a network page
US9183280B2 (en) * 2011-09-30 2015-11-10 Paypal, Inc. Methods and systems using demand metrics for presenting aspects for item listings presented in a search results page
US20130089847A1 (en) * 2011-10-11 2013-04-11 John Allen Baker SYSTEMS AND METHODS FOR MONITORING eLEARNING SYSTEM DATA AND GENERATING RECOMMENDATIONS
US20130159254A1 (en) * 2011-12-14 2013-06-20 Yahoo! Inc. System and methods for providing content via the internet
US20130166379A1 (en) * 2011-12-21 2013-06-27 Akintunde Ehindero Social Targeting
US20130166609A1 (en) * 2011-12-22 2013-06-27 General Instrument Corporation Hierarchical behavioral profile
US8943015B2 (en) 2011-12-22 2015-01-27 Google Technology Holdings LLC Hierarchical behavioral profile
US9110998B2 (en) 2011-12-22 2015-08-18 Google Technology Holdings LLC Hierarchical behavioral profile
CN102402766B (en) * 2011-12-27 2017-07-28 北京京东尚科信息技术有限公司 A user interest modeling method based web browser
JP5668007B2 (en) * 2012-03-22 2015-02-12 日本電信電話株式会社 Interest analysis method, apparatus and program
US20130275353A1 (en) * 2012-04-16 2013-10-17 Apple Inc. Systems and methods for selecting media items
US9053185B1 (en) 2012-04-30 2015-06-09 Google Inc. Generating a representative model for a plurality of models identified by similar feature data
US9465875B2 (en) 2012-05-31 2016-10-11 Hewlett Packard Enterprise Development Lp Searching based on an identifier of a searcher
US9043722B1 (en) 2012-06-19 2015-05-26 Surfwax, Inc. User interfaces for displaying relationships between cells in a grid
US20140019443A1 (en) * 2012-07-10 2014-01-16 Venor, Inc. Systems and methods for discovering content of predicted interest to a user
US9336302B1 (en) 2012-07-20 2016-05-10 Zuci Realty Llc Insight and algorithmic clustering for automated synthesis
US9588784B2 (en) * 2012-08-28 2017-03-07 Yahoo! Inc. System and method for providing selective and visible personalized content for webpages
US9065727B1 (en) 2012-08-31 2015-06-23 Google Inc. Device identifier similarity models derived from online event signals
US20150242751A1 (en) * 2012-09-17 2015-08-27 New York University System and method for estimating audience interest
US9852215B1 (en) * 2012-09-21 2017-12-26 Amazon Technologies, Inc. Identifying text predicted to be of interest
US9015088B2 (en) * 2012-10-30 2015-04-21 Palo Alto Research Center Incorported Method and system for psychological analysis by fusing multiple-view predictions
US9278255B2 (en) 2012-12-09 2016-03-08 Arris Enterprises, Inc. System and method for activity recognition
US9348886B2 (en) * 2012-12-19 2016-05-24 Facebook, Inc. Formation and description of user subgroups
US9542669B1 (en) * 2013-03-14 2017-01-10 Blab, Inc. Encoding and using information about distributed group discussions
US9477785B2 (en) * 2013-03-15 2016-10-25 NutraSpace LLC Customized query application and data result updating procedure
US9430572B2 (en) * 2013-03-15 2016-08-30 Excalibur Ip, Llc Method and system for user profiling via mapping third party interests to a universal interest space
US9547698B2 (en) 2013-04-23 2017-01-17 Google Inc. Determining media consumption preferences
US9244522B2 (en) * 2013-08-30 2016-01-26 Linkedin Corporation Guided browsing experience
US9542440B2 (en) 2013-11-04 2017-01-10 Microsoft Technology Licensing, Llc Enterprise graph search based on object and actor relationships
US20150149261A1 (en) * 2013-11-25 2015-05-28 Flipboard, Inc. Measuring quality of content items presented by a digital magazine server
US20150169772A1 (en) * 2013-12-12 2015-06-18 Microsoft Corporation Personalizing Search Results Based on User-Generated Content
US9208153B1 (en) * 2013-12-13 2015-12-08 Symantec Corporation Filtering relevant event notifications in a file sharing and collaboration environment
US20150178807A1 (en) 2013-12-20 2015-06-25 Zumur, LLC System and method for asynchronous consumer item searching requests with synchronous parallel searching
US9342586B2 (en) * 2014-02-03 2016-05-17 International Business Machines Corporation Managing and using shareable search lists
US9870432B2 (en) 2014-02-24 2018-01-16 Microsoft Technology Licensing, Llc Persisted enterprise graph queries
US9959317B2 (en) 2014-03-05 2018-05-01 Empire Technology Development Llc Semantic data generation
US20160259765A1 (en) * 2014-05-04 2016-09-08 Daqi Li Document interaction evaluator based on an ontology
US20150356174A1 (en) * 2014-06-06 2015-12-10 Wipro Limited System and methods for capturing and analyzing documents to identify ideas in the documents
US20150371677A1 (en) * 2014-06-19 2015-12-24 BrightSky Labs, Inc. User interface for video editing system
CN104199836B (en) * 2014-08-04 2017-07-14 浙江工商大学 One kind of labeling method for constructing user model based on sub-division of interest
CN104281696B (en) * 2014-10-16 2017-09-15 江西师范大学 An active space for personalized information distribution method
US20160171382A1 (en) * 2014-12-16 2016-06-16 Facebook, Inc. Systems and methods for page recommendations based on online user behavior
JP2016162163A (en) * 2015-03-02 2016-09-05 富士ゼロックス株式会社 Information processor and information processing program
CN105095187A (en) * 2015-08-07 2015-11-25 广州神马移动信息科技有限公司 Search intention identification method and device
US20170124184A1 (en) * 2015-10-29 2017-05-04 Intuit Inc. Method and system for performing a probabilistic topic analysis of search queries for a customer support system
US20170169341A1 (en) * 2015-12-14 2017-06-15 Le Holdings (Beijing) Co., Ltd. Method for intelligent recommendation

Citations (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5704017A (en) 1996-02-16 1997-12-30 Microsoft Corporation Collaborative filtering utilizing a belief network
US5754939A (en) * 1994-11-29 1998-05-19 Herz; Frederick S. M. System for generation of user profiles for a system for customized electronic identification of desirable objects
US5867799A (en) 1996-04-04 1999-02-02 Lang; Andrew K. Information system and method for filtering a massive flow of information entities to meet user information classification needs
US5918014A (en) 1995-12-27 1999-06-29 Athenium, L.L.C. Automated collaborative filtering in world wide web advertising
US5933827A (en) 1996-09-25 1999-08-03 International Business Machines Corporation System for identifying new web pages of interest to a user
US5964839A (en) * 1996-03-29 1999-10-12 At&T Corp System and method for monitoring information flow and performing data collection
US5991735A (en) * 1996-04-26 1999-11-23 Be Free, Inc. Computer program apparatus for determining behavioral profile of a computer user
US5999975A (en) 1997-03-28 1999-12-07 Nippon Telegraph And Telephone Corporation On-line information providing scheme featuring function to dynamically account for user's interest
US6006218A (en) * 1997-02-28 1999-12-21 Microsoft Methods and apparatus for retrieving and/or processing retrieved information as a function of a user's estimated knowledge
US6041311A (en) 1995-06-30 2000-03-21 Microsoft Corporation Method and apparatus for item recommendation using automated collaborative filtering
US6564170B2 (en) * 2000-12-29 2003-05-13 Hewlett-Packard Development Company, L.P. Customizable user interfaces
US6567850B1 (en) * 1998-10-28 2003-05-20 Yodlee, Inc. System and method for determining revenue from an intermediary derived from servicing data requests
US6687696B2 (en) 2000-07-26 2004-02-03 Recommind Inc. System and method for personalized search, information filtering, and for generating recommendations utilizing statistical latent class models
US6732090B2 (en) * 2001-08-13 2004-05-04 Xerox Corporation Meta-document management system with user definable personalities

Family Cites Families (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6460036B1 (en) * 1994-11-29 2002-10-01 Pinpoint Incorporated System and method for providing customized electronic newspapers and target advertisements
US6029195A (en) * 1994-11-29 2000-02-22 Herz; Frederick S. M. System for customized electronic identification of desirable objects
DE19529277A1 (en) * 1995-08-09 1997-02-13 Bayerische Motoren Werke Ag A method of operating a hydraulically-controlled / regulated camshaft adjusting device for internal combustion engines
US6182068B1 (en) * 1997-08-01 2001-01-30 Ask Jeeves, Inc. Personalized search methods
US6345293B1 (en) * 1997-07-03 2002-02-05 Microsoft Corporation Personalized information for an end user transmitted over a computer network
US6230204B1 (en) * 1997-12-19 2001-05-08 Micron Electronics, Inc. Method and system for estimating usage of computer resources
EP1051683B1 (en) * 1998-01-30 2003-07-23 Easynet Access Inc. Personalized internet interaction
US6182133B1 (en) * 1998-02-06 2001-01-30 Microsoft Corporation Method and apparatus for display of information prefetching and cache status having variable visual indication based on a period of time since prefetching
US6582475B2 (en) * 1998-09-09 2003-06-24 Ricoh Company Limited Automatic adaptive document printing help system
US6567797B1 (en) * 1999-01-26 2003-05-20 Xerox Corporation System and method for providing recommendations based on multi-modal user clusters
US7353246B1 (en) * 1999-07-30 2008-04-01 Miva Direct, Inc. System and method for enabling information associations
WO2001014952A3 (en) * 1999-08-26 2001-06-14 Memetrics Inc On-line experimentation
US6697824B1 (en) * 1999-08-31 2004-02-24 Accenture Llp Relationship management in an E-commerce application framework
US6828992B1 (en) * 1999-11-04 2004-12-07 Koninklijke Philips Electronics N.V. User interface with dynamic menu option organization
US6981040B1 (en) * 1999-12-28 2005-12-27 Utopy, Inc. Automatic, personalized online information and product services
GB0003411D0 (en) * 2000-02-15 2000-04-05 Dialog Corp The Plc Accessing data
US6606620B1 (en) * 2000-07-24 2003-08-12 International Business Machines Corporation Method and system for classifying semi-structured documents
US6874019B2 (en) * 2001-03-08 2005-03-29 International Business Machines Corporation Predictive caching and highlighting of web pages
US6915482B2 (en) * 2001-03-28 2005-07-05 Cyber Watcher As Method and arrangement for web information monitoring
US7149704B2 (en) * 2001-06-29 2006-12-12 Claria Corporation System, method and computer program product for collecting information about a network user
JP2003085204A (en) * 2001-09-11 2003-03-20 Ricoh Co Ltd Document processing managing device
US20060235938A1 (en) * 2002-11-12 2006-10-19 Pennell Mark E System and method for delivery of information based on web page content

Patent Citations (16)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5754939A (en) * 1994-11-29 1998-05-19 Herz; Frederick S. M. System for generation of user profiles for a system for customized electronic identification of desirable objects
US6041311A (en) 1995-06-30 2000-03-21 Microsoft Corporation Method and apparatus for item recommendation using automated collaborative filtering
US5918014A (en) 1995-12-27 1999-06-29 Athenium, L.L.C. Automated collaborative filtering in world wide web advertising
US5704017A (en) 1996-02-16 1997-12-30 Microsoft Corporation Collaborative filtering utilizing a belief network
US5964839A (en) * 1996-03-29 1999-10-12 At&T Corp System and method for monitoring information flow and performing data collection
US6029161A (en) 1996-04-04 2000-02-22 Lycos, Inc. Multi-level mindpool system especially adapted to provide collaborative filter data for a large scale information filtering system
US5983214A (en) 1996-04-04 1999-11-09 Lycos, Inc. System and method employing individual user content-based data and user collaborative feedback data to evaluate the content of an information entity in a large information communication network
US5867799A (en) 1996-04-04 1999-02-02 Lang; Andrew K. Information system and method for filtering a massive flow of information entities to meet user information classification needs
US5991735A (en) * 1996-04-26 1999-11-23 Be Free, Inc. Computer program apparatus for determining behavioral profile of a computer user
US5933827A (en) 1996-09-25 1999-08-03 International Business Machines Corporation System for identifying new web pages of interest to a user
US6006218A (en) * 1997-02-28 1999-12-21 Microsoft Methods and apparatus for retrieving and/or processing retrieved information as a function of a user's estimated knowledge
US5999975A (en) 1997-03-28 1999-12-07 Nippon Telegraph And Telephone Corporation On-line information providing scheme featuring function to dynamically account for user's interest
US6567850B1 (en) * 1998-10-28 2003-05-20 Yodlee, Inc. System and method for determining revenue from an intermediary derived from servicing data requests
US6687696B2 (en) 2000-07-26 2004-02-03 Recommind Inc. System and method for personalized search, information filtering, and for generating recommendations utilizing statistical latent class models
US6564170B2 (en) * 2000-12-29 2003-05-13 Hewlett-Packard Development Company, L.P. Customizable user interfaces
US6732090B2 (en) * 2001-08-13 2004-05-04 Xerox Corporation Meta-document management system with user definable personalities

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
Basset et al. (2002) in a paper entitled "A Study of Generalization Techniques in Evolutionary Rule Learning".
Mobasher, B., Automatic personalization based on web usage mining, http://maya.cs.depaul.edu/~mobasher/personalization.
Yann LeCun (2004) in a lecture entitled "Machine Learning and Pattern Recognition" and presented at The Courant Institute, New York University.

Cited By (631)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8755763B2 (en) 1998-01-22 2014-06-17 Black Hills Media Method and device for an internet radio capable of obtaining playlist content from a content server
US8918480B2 (en) 1998-01-22 2014-12-23 Black Hills Media, Llc Method, system, and device for the distribution of internet radio content
US8792850B2 (en) 1998-01-22 2014-07-29 Black Hills Media Method and device for obtaining playlist content over a network
US9312827B2 (en) 1998-01-22 2016-04-12 Black Hills Media, Llc Network enabled audio device and radio site
US9397627B2 (en) 1998-01-22 2016-07-19 Black Hills Media, Llc Network-enabled audio device
US8352984B2 (en) 1998-06-12 2013-01-08 Thomson Licensing System and method for generating and managing user preference information for scheduled and stored television programs
US8351763B2 (en) 1998-06-12 2013-01-08 Thomson Licensing Television program recording with user preference determination
US20130182835A1 (en) * 1999-07-09 2013-07-18 Jpmorgan Chase Bank, N.A. System and Method of Intelligent Call Routing for Cross Sell Offer Selection based on Optimization Parameters or Account-level Data
US20140133646A1 (en) * 1999-07-09 2014-05-15 Jpmorgan Chase Bank, N.A. System and method of intelligent call routing for cross sell offer selection based on optimization parameters or account-level data
US8687790B2 (en) * 1999-07-09 2014-04-01 Jpmorgan Chase Bank, N.A. System and method of intelligent call routing for cross sell offer selection based on optimization parameters or account-level data
US8995643B2 (en) * 1999-07-09 2015-03-31 Jpmorgan Chase Bank, N.A. System and method of intelligent call routing for cross sell offer selection based on optimization parameters or account-level data
US7353246B1 (en) * 1999-07-30 2008-04-01 Miva Direct, Inc. System and method for enabling information associations
US20080114756A1 (en) * 1999-12-28 2008-05-15 Levino Authomatic, personalized online information and product services
US7685276B2 (en) 1999-12-28 2010-03-23 Personalized User Model Automatic, personalized online information and product services
US20100122178A1 (en) * 1999-12-28 2010-05-13 Personalized User Model Automatic, personalized online information and product services
US7320031B2 (en) * 1999-12-28 2008-01-15 Utopy, Inc. Automatic, personalized online information and product services
US20060136589A1 (en) * 1999-12-28 2006-06-22 Utopy, Inc. Automatic, personalized online information and product services
US8825853B1 (en) * 1999-12-28 2014-09-02 Personalized User Model Automatic, personalized online information and product services
US9098813B1 (en) * 1999-12-28 2015-08-04 Personalized User Model Methods and systems for conducting a personalized web browser search by probability estimation that is unseen document is of interest to the user
US8626911B2 (en) 1999-12-28 2014-01-07 Personalized User Model Automatic, personalized online information and product services
US8301764B2 (en) 1999-12-28 2012-10-30 Personalized User Model Method and system for personalized searching of information and product services by estimating an interest to a user
USRE46536E1 (en) 2000-04-14 2017-09-05 Intel Corporation System and method of managing metadata data
USRE42101E1 (en) * 2000-04-14 2011-02-01 Realnetworks, Inc. System and method of managing metadata data
US8087051B2 (en) 2000-06-30 2011-12-27 Thomson Licensing Database management system and method for electronic program guide and television channel lineup organization
US8712834B2 (en) 2000-07-18 2014-04-29 Google Inc. Economic filtering system for delivery of permission based, targeted, incentivized advertising
US20100280906A1 (en) * 2000-07-18 2010-11-04 Lim Stanley T Economic Filtering System for Delivery of Permission Based, Targeted, Incentivized Advertising
US7558773B2 (en) * 2000-07-19 2009-07-07 Convergys Cmg Utah, Inc. Expert supported interactive product selection and recommendation
US20070208682A1 (en) * 2000-07-19 2007-09-06 Convergys Cmg Utah, Inc. Expert supported interactive product selection and recommendation
US20020040386A1 (en) * 2000-08-31 2002-04-04 Bertrand Berthelot Method and device for adapting the content of documents of an information server
US9268775B1 (en) 2000-09-07 2016-02-23 Black Hills Media, Llc Method and system for providing an audio element cache in a customized personal radio broadcast
US7840691B1 (en) 2000-09-07 2010-11-23 Zamora Radio, Llc Personal broadcast server system for providing a customized broadcast
US8667161B2 (en) 2000-09-07 2014-03-04 Black Hills Media Personal broadcast server system for providing a customized broadcast
US8302127B2 (en) 2000-09-25 2012-10-30 Thomson Licensing System and method for personalized TV
US8060816B1 (en) * 2000-10-31 2011-11-15 International Business Machines Corporation Methods and apparatus for intelligent crawling on the world wide web
US8230041B2 (en) * 2000-11-07 2012-07-24 Ricoh Company, Ltd. System, method, apparatus, and program for providing electronic manual
US20070033230A1 (en) * 2000-11-07 2007-02-08 Kiyohisa Minowa System, method, apparatus, and program for providing electronic manual
US20020069223A1 (en) * 2000-11-17 2002-06-06 Goodisman Aaron A. Methods and systems to link data
US8850465B2 (en) 2001-01-09 2014-09-30 Thomson Licensing System, method, and software application for targeted advertising via behavioral model clustering, and preference programming based on behavioral model clusters
US9277264B2 (en) 2001-01-09 2016-03-01 Thomson Licensing System, method, and software application for targeted advertising via behavioral model clustering, and preference programming based on behavioral model clusters
US8495680B2 (en) * 2001-01-09 2013-07-23 Thomson Licensing System and method for behavioral model clustering in television usage, targeted advertising via model clustering, and preference programming based on behavioral model clusters
US20030101451A1 (en) * 2001-01-09 2003-05-29 Isaac Bentolila System, method, and software application for targeted advertising via behavioral model clustering, and preference programming based on behavioral model clusters
US8046797B2 (en) 2001-01-09 2011-10-25 Thomson Licensing System, method, and software application for targeted advertising via behavioral model clustering, and preference programming based on behavioral model clusters
US20030101449A1 (en) * 2001-01-09 2003-05-29 Isaac Bentolila System and method for behavioral model clustering in television usage, targeted advertising via model clustering, and preference programming based on behavioral model clusters
US20060246932A1 (en) * 2001-01-16 2006-11-02 Texas Instruments Incorporated Collaborative Mechanism of Enhanced Coexistence of Collocated Wireless Networks
US9047386B2 (en) 2001-03-07 2015-06-02 Bascom Research, Llc Creating a link relationship between document objects and storing in a link directory
US9128934B2 (en) 2001-03-07 2015-09-08 Bascom Research, Llc User interface for presenting and searching relationships between document objects located on a network
US7389241B1 (en) * 2001-03-07 2008-06-17 Thomas Layne Bascom Method for users of a network to provide other users with access to link relationships between documents
US8515998B1 (en) 2001-03-07 2013-08-20 Bascom Research LLP Framework for managing document objects stored on a network
US9218431B2 (en) 2001-03-07 2015-12-22 Bascom Research, Llc System for linked and networked document objects
US9043352B1 (en) 2001-03-07 2015-05-26 Bascom Research, Llc Method for searching document objects on a network
US20030014399A1 (en) * 2001-03-12 2003-01-16 Hansen Mark H. Method for organizing records of database search activity by topical relevance
US7194454B2 (en) * 2001-03-12 2007-03-20 Lucent Technologies Method for organizing records of database search activity by topical relevance
US20040210541A1 (en) * 2001-05-04 2004-10-21 Jeremy Epstien User interface for a rules engine and methods therefor
US20070083343A1 (en) * 2001-06-08 2007-04-12 Panoratio Database Images Gmbh Statistical models for improving the performance of database operations
US20020198962A1 (en) * 2001-06-21 2002-12-26 Horn Frederic A. Method, system, and computer program product for distributing a stored URL and web document set
US20060218245A1 (en) * 2001-06-21 2006-09-28 Horn Frederic A Method, system, and computer program product for distributing a stored URL and web document set
US8538892B2 (en) 2001-06-21 2013-09-17 Thomson Licenging Method and system for electronic purchases using an intelligent data carrier medium, electronic coupon system, and interactive TV infrastructure
US8818871B2 (en) 2001-06-21 2014-08-26 Thomson Licensing Method and system for electronic purchases using an intelligent data carrier medium, electronic coupon system, and interactive TV infrastructure
US20130054410A1 (en) * 2001-07-17 2013-02-28 Incucomm, Incorporated System and Method for Providing Requested Information to Thin Clients
US7860925B1 (en) 2001-10-19 2010-12-28 Outlooksoft Corporation System and method for adaptively selecting and delivering recommendations to a requester
US7231419B1 (en) * 2001-10-19 2007-06-12 Outlooksoft Corporation System and method for adaptively selecting and delivering recommendations to a requester
US7120647B2 (en) * 2001-10-30 2006-10-10 General Electric Company Web-based method and system for providing expert information on selected matters
US20030101083A1 (en) * 2001-10-30 2003-05-29 Sambedu Venkatesh Web-based method and system for providing expert information on selected matters
US7716256B2 (en) * 2001-11-19 2010-05-11 Fujitsu Limited Information navigation system
US20040210558A1 (en) * 2001-11-19 2004-10-21 Fujitsu Limited Information navigation system
US20060009273A2 (en) * 2002-02-14 2006-01-12 Waterleaf Limited Menu Selection System and Method of Operation Thereof
US20050159204A1 (en) * 2002-02-14 2005-07-21 Martin Moshal Menu selection system and method of operation thereof
US20090240769A1 (en) * 2002-02-25 2009-09-24 Ascentive Llc Method and system for screening remote site connections and filtering data based on a community trust assessment
US8560596B2 (en) * 2002-02-25 2013-10-15 Ascentive Llc Method and system for screening remote site connections and filtering data based on a community trust assessment
US20030172143A1 (en) * 2002-03-06 2003-09-11 Koji Wakayama Access node apparatus and method for internet using condition analysis
US7814139B2 (en) 2002-03-07 2010-10-12 Complete, Inc. Systems and methods for clickstream analysis to modify an off-line business process involving forecasting demand
US7797371B2 (en) 2002-03-07 2010-09-14 Compete, Inc. Systems and methods for clickstream analysis to modify an off-line business process involving determining related or complementary items
US20080183796A1 (en) * 2002-03-07 2008-07-31 Man Jit Singh Clickstream analysis methods and systems
US9092788B2 (en) 2002-03-07 2015-07-28 Compete, Inc. System and method of collecting and analyzing clickstream data
US7979544B2 (en) 2002-03-07 2011-07-12 Compete, Inc. Computer program product and method for estimating internet traffic
US9123056B2 (en) 2002-03-07 2015-09-01 Compete, Inc. Clickstream analysis methods and systems related to modifying an offline promotion for a consumer good
US9129032B2 (en) 2002-03-07 2015-09-08 Compete, Inc. System and method for processing a clickstream in a parallel processing architecture
US8626834B2 (en) 2002-03-07 2014-01-07 Compete, Inc. Clickstream analysis methods and systems related to modifying an offline promotion for a consumer good
US8055709B2 (en) 2002-03-07 2011-11-08 Compete, Inc. Systems and methods for clickstream analysis to modify an off-line business process involving product pricing
US8095589B2 (en) * 2002-03-07 2012-01-10 Compete, Inc. Clickstream analysis methods and systems
US8095621B2 (en) 2002-03-07 2012-01-10 Compete, Inc. Systems and methods for clickstream analysis to modify an off-line business process involving automobile sales
US8099496B2 (en) 2002-03-07 2012-01-17 Compete, Inc. Systems and methods for clickstream analysis to modify an off-line business process involving matching a distribution list
US8135833B2 (en) 2002-03-07 2012-03-13 Compete, Inc. Computer program product and method for estimating internet traffic
US9292860B2 (en) 2002-03-07 2016-03-22 Compete, Inc. Clickstream analysis methods and systems related to modifying an offline promotion for a consumer good
US20030171977A1 (en) * 2002-03-07 2003-09-11 Compete, Inc. Clickstream analysis methods and systems
US7895258B2 (en) 2002-03-07 2011-02-22 Compete, Inc. Systems and methods for clickstream analysis to modify an off-line business process involving matching a sales medium
US9501781B2 (en) 2002-03-07 2016-11-22 Comscore, Inc. Clickstream analysis methods and systems related to improvements in online stores and media content
US20080183718A1 (en) * 2002-03-07 2008-07-31 Man Jit Singh Clickstream analysis methods and systems
US8769080B2 (en) 2002-03-07 2014-07-01 Compete, Inc. System and method for a behavior-targeted survey
US20080183867A1 (en) * 2002-03-07 2008-07-31 Man Jit Singh Clickstream analysis methods and systems
US20080183868A1 (en) * 2002-03-07 2008-07-31 Man Jit Singh Clickstream analysis methods and systems
US20080177779A1 (en) * 2002-03-07 2008-07-24 David Cancel Presentation of media segments
US20080177778A1 (en) * 2002-03-07 2008-07-24 David Cancel Presentation of media segments
US20080183806A1 (en) * 2002-03-07 2008-07-31 David Cancel Presentation of media segments
US20080183805A1 (en) * 2002-03-07 2008-07-31 David Cancel Presentation of media segments
US20080183717A1 (en) * 2002-03-07 2008-07-31 Man Jit Singh Clickstream analysis methods and systems
US20100030894A1 (en) * 2002-03-07 2010-02-04 David Cancel Computer program product and method for estimating internet traffic
US8356097B2 (en) 2002-03-07 2013-01-15 Compete, Inc. Computer program product and method for estimating internet traffic
US7043521B2 (en) * 2002-03-21 2006-05-09 Rockwell Electronic Commerce Technologies, Llc Search agent for searching the internet
US20030217097A1 (en) * 2002-03-21 2003-11-20 Eitel Robert T. Search agent for searching the internet
US9684676B1 (en) 2002-03-29 2017-06-20 Google Inc. Method for searching media
USRE45952E1 (en) 2002-03-29 2016-03-29 Google Inc. Method for searching media
US9015159B1 (en) * 2002-03-29 2015-04-21 Google Inc. Method for searching media
US7574427B2 (en) * 2002-05-17 2009-08-11 Delefevre Patrick Y Neutral sales consultant
US20070011131A1 (en) * 2002-05-17 2007-01-11 Delefevre Patrick Y Neutral sales consultant
US7890451B2 (en) 2002-10-09 2011-02-15 Compete, Inc. Computer program product and method for refining an estimate of internet traffic
US20060026642A1 (en) * 2002-12-11 2006-02-02 Koninklijke Philips Electronics, N.V. Method and apparatus for predicting a number of individuals interested in an item based on recommendations of such item
US20040143667A1 (en) * 2003-01-17 2004-07-22 Jason Jerome Content distribution system
US8583764B2 (en) * 2003-01-17 2013-11-12 Relevant Media Llc Content distribution system
US20100070340A1 (en) * 2003-01-17 2010-03-18 Relevant Media Llc Content distribution system
US7526545B2 (en) * 2003-01-17 2009-04-28 Relevant Media Llc Content distribution system
US20060195583A1 (en) * 2003-02-27 2006-08-31 Fabio Bellifemine Method and system for providing information services to a client using a user profile
US20100306079A1 (en) * 2003-03-31 2010-12-02 Sap Ag Collaborative Product Taxonomy Instantiation
US8498906B2 (en) * 2003-03-31 2013-07-30 Sap Ag Collaborative product taxonomy instantiation
US20110173181A1 (en) * 2003-04-24 2011-07-14 Chang William I Search engine and method with improved relevancy, scope, and timeliness
US20050004943A1 (en) * 2003-04-24 2005-01-06 Chang William I. Search engine and method with improved relevancy, scope, and timeliness
US8886621B2 (en) 2003-04-24 2014-11-11 Affini, Inc. Search engine and method with improved relevancy, scope, and timeliness
US8645345B2 (en) 2003-04-24 2014-02-04 Affini, Inc. Search engine and method with improved relevancy, scope, and timeliness
US7917483B2 (en) * 2003-04-24 2011-03-29 Affini, Inc. Search engine and method with improved relevancy, scope, and timeliness
US8799763B1 (en) 2003-04-29 2014-08-05 Google Inc. Methods and apparatus for assisting in completion of a form
US8190989B1 (en) 2003-04-29 2012-05-29 Google Inc. Methods and apparatus for assisting in completion of a form
US20050010951A1 (en) * 2003-05-14 2005-01-13 Sony Corporation Information processing apparatus and method, program, and recording medium
US7487164B2 (en) * 2003-05-14 2009-02-03 Sony Corporation Information processing apparatus capable of properly reflecting a change in a user's preference or interest
US20050038638A1 (en) * 2003-07-25 2005-02-17 Dorin Comaniciu Density morphing and mode propagation for Bayesian filtering
US7526414B2 (en) * 2003-07-25 2009-04-28 Siemens Corporate Research, Inc. Density morphing and mode propagation for Bayesian filtering
US20050071328A1 (en) * 2003-09-30 2005-03-31 Lawrence Stephen R. Personalization of web search
US20130282713A1 (en) * 2003-09-30 2013-10-24 Stephen R. Lawrence Personalization of Web Search Results Using Term, Category, and Link-Based User Profiles
US20160342695A1 (en) * 2003-09-30 2016-11-24 Google Inc. Personalization of web search results using term, category, and link-based user profiles
US9298777B2 (en) * 2003-09-30 2016-03-29 Google Inc. Personalization of web search results using term, category, and link-based user profiles
US20100228715A1 (en) * 2003-09-30 2010-09-09 Lawrence Stephen R Personalization of Web Search Results Using Term, Category, and Link-Based User Profiles
US20120233142A1 (en) * 2003-09-30 2012-09-13 Lawrence Stephen R Personalization of Web Search Results Using Term, Category, and Link-Based User Profiles
US9697249B1 (en) 2003-09-30 2017-07-04 Google Inc. Estimating confidence for query revision models
US8024323B1 (en) * 2003-11-13 2011-09-20 AudienceScience Inc. Natural language search for audience
US20050114382A1 (en) * 2003-11-26 2005-05-26 Lakshminarayan Choudur K. Method and system for data segmentation
USRE44968E1 (en) 2003-11-28 2014-06-24 World Assets Consulting Ag, Llc Adaptive self-modifying and recombinant systems
USRE44966E1 (en) 2003-11-28 2014-06-24 World Assets Consulting Ag, Llc Adaptive recommendations systems
USRE44967E1 (en) 2003-11-28 2014-06-24 World Assets Consulting Ag, Llc Adaptive social and process network systems
US8566263B2 (en) 2003-11-28 2013-10-22 World Assets Consulting Ag, Llc Adaptive computer-based personalities
US20070156614A1 (en) * 2003-11-28 2007-07-05 Manyworlds, Inc. Adaptive Fuzzy Network System and Method
USRE45770E1 (en) 2003-11-28 2015-10-20 World Assets Consulting Ag, Llc Adaptive recommendation explanations
USRE44559E1 (en) 2003-11-28 2013-10-22 World Assets Consulting Ag, Llc Adaptive social computing methods
US8600920B2 (en) 2003-11-28 2013-12-03 World Assets Consulting Ag, Llc Affinity propagation in adaptive network-based systems
US8612415B2 (en) 2003-12-03 2013-12-17 Google Inc. Personalized network searching
US8886626B2 (en) 2003-12-03 2014-11-11 Google Inc. Personalized network searching
US8015170B2 (en) * 2003-12-03 2011-09-06 Google Inc. Personalized network searching
US8150830B2 (en) 2003-12-03 2012-04-03 Google Inc. Personalized network searching
US20080215553A1 (en) * 2003-12-03 2008-09-04 Google Inc. Personalized Network Searching
US8166017B2 (en) 2003-12-03 2012-04-24 Google Inc. Personalized network searching
US20050131866A1 (en) * 2003-12-03 2005-06-16 Badros Gregory J. Methods and systems for personalized network searching
US9679067B2 (en) 2003-12-03 2017-06-13 Google Inc. Personalized network searching
US7523096B2 (en) * 2003-12-03 2009-04-21 Google Inc. Methods and systems for personalized network searching
US7890584B2 (en) * 2003-12-10 2011-02-15 International Business Machines Corporation System for information capture
US20090083388A1 (en) * 2003-12-10 2009-03-26 International Business Machines Corporation System for Information Capture
US8554693B2 (en) 2003-12-17 2013-10-08 At&T Intellectual Property I, L.P. Methods, systems and storage medium for providing content notification based on user communities
US20050137886A1 (en) * 2003-12-17 2005-06-23 Kim Littrell Methods, systems and storage medium for providing content notification based on user communities
US20110066958A1 (en) * 2003-12-17 2011-03-17 At&T Intellectual Property I, L.P. Via Transfer From Bellsouth Intellectual Property Corporation Methods, Systems And Storage Medium For Providing Content Notification Based On User Communities
US7860804B2 (en) 2003-12-17 2010-12-28 At&T Intellectual Property, I, L.P. Methods, systems and storage medium for providing content notification based on user communities
US20050149619A1 (en) * 2003-12-22 2005-07-07 Cowan Keith O. Methods, systems and storage medium for distributing content based on user communities
US20080016101A1 (en) * 2003-12-30 2008-01-17 Shopping.Com Systems and methods for dynamically updating relevance of a selected item
US8655905B2 (en) * 2003-12-30 2014-02-18 Microsoft Corporation Incremental query refinement
US8065301B2 (en) * 2003-12-30 2011-11-22 Shopping.Com Systems and methods for dynamically updating relevance of a selected item
US20120136886A1 (en) * 2003-12-30 2012-05-31 Microsoft Corporation Incremental query refinement
US9245052B2 (en) 2003-12-30 2016-01-26 Microsoft Technology Licensing, Llc Incremental query refinement
US20050149498A1 (en) * 2003-12-31 2005-07-07 Stephen Lawrence Methods and systems for improving a search ranking using article information
US8954420B1 (en) 2003-12-31 2015-02-10 Google Inc. Methods and systems for improving a search ranking using article information
US8676837B2 (en) 2003-12-31 2014-03-18 Google Inc. Systems and methods for personalizing aggregated news content
US20050144084A1 (en) * 2003-12-31 2005-06-30 Gold Kevin K. Using a customer's purchasing intent in recommending alternative items for purchase in a virtual store
US8156444B1 (en) 2003-12-31 2012-04-10 Google Inc. Systems and methods for determining a user interface attribute
US9235849B2 (en) * 2003-12-31 2016-01-12 Google Inc. Generating user information for use in targeted advertising
US20050131762A1 (en) * 2003-12-31 2005-06-16 Krishna Bharat Generating user information for use in targeted advertising
US8126865B1 (en) * 2003-12-31 2012-02-28 Google Inc. Systems and methods for syndicating and hosting customized news content
US8832058B1 (en) 2003-12-31 2014-09-09 Google Inc. Systems and methods for syndicating and hosting customized news content
US7392278B2 (en) * 2004-01-23 2008-06-24 Microsoft Corporation Building and using subwebs for focused search
US7493323B2 (en) * 2004-02-27 2009-02-17 Ricoh Company, Ltd. Document group analyzing apparatus, a document group analyzing method, a document group analyzing system, a program, and a recording medium
US20050192949A1 (en) * 2004-02-27 2005-09-01 Yuichi Kojima Document group analyzing apparatus, a document group analyzing method, a document group analyzing system, a program, and a recording medium
US8874567B2 (en) 2004-03-29 2014-10-28 Google Inc. Variable personalization of search results in a search engine
US9058364B2 (en) 2004-03-29 2015-06-16 Google Inc. Variable personalization of search results in a search engine
US8812515B1 (en) 2004-03-31 2014-08-19 Google Inc. Processing contact information
US8099407B2 (en) 2004-03-31 2012-01-17 Google Inc. Methods and systems for processing media files
US7680809B2 (en) 2004-03-31 2010-03-16 Google Inc. Profile based capture component
US7680888B1 (en) 2004-03-31 2010-03-16 Google Inc. Methods and systems for processing instant messenger messages
US7412708B1 (en) 2004-03-31 2008-08-12 Google Inc. Methods and systems for capturing information
US8346777B1 (en) 2004-03-31 2013-01-01 Google Inc. Systems and methods for selectively storing event data
US7333976B1 (en) 2004-03-31 2008-02-19 Google Inc. Methods and systems for processing contact information
US7725508B2 (en) 2004-03-31 2010-05-25 Google Inc. Methods and systems for information capture and retrieval
US8386728B1 (en) 2004-03-31 2013-02-26 Google Inc. Methods and systems for prioritizing a crawl
US7941439B1 (en) 2004-03-31 2011-05-10 Google Inc. Methods and systems for information capture
US8631076B1 (en) 2004-03-31 2014-01-14 Google Inc. Methods and systems for associating instant messenger events
US9836544B2 (en) 2004-03-31 2017-12-05 Google Inc. Methods and systems for prioritizing a crawl
US8275839B2 (en) 2004-03-31 2012-09-25 Google Inc. Methods and systems for processing email messages
US8595214B1 (en) 2004-03-31 2013-11-26 Google Inc. Systems and methods for article location and retrieval
US7581227B1 (en) 2004-03-31 2009-08-25 Google Inc. Systems and methods of synchronizing indexes
US9189553B2 (en) 2004-03-31 2015-11-17 Google Inc. Methods and systems for prioritizing a crawl
US9311408B2 (en) 2004-03-31 2016-04-12 Google, Inc. Methods and systems for processing media files
US8161053B1 (en) 2004-03-31 2012-04-17 Google Inc. Methods and systems for eliminating duplicate events
US20080059419A1 (en) * 2004-03-31 2008-03-06 David Benjamin Auerbach Systems and methods for providing search results
US20050228642A1 (en) * 2004-04-06 2005-10-13 Microsoft Corporation Efficient capitalization through user modeling
US7827025B2 (en) * 2004-04-06 2010-11-02 Microsoft Corporation Efficient capitalization through user modeling
US9554405B2 (en) 2004-05-05 2017-01-24 Black Hills Media, Llc Wireless speaker for receiving from a mobile phone directions to receive and render a playlist from a content server on the internet
US9516370B1 (en) 2004-05-05 2016-12-06 Black Hills Media, Llc Method, device, and system for directing a wireless speaker from a mobile phone to receive and render a playlist from a content server on the internet
US20110153452A1 (en) * 2004-05-20 2011-06-23 Manyworlds, Inc. Contextual Commerce Systems and Methods
USRE43768E1 (en) 2004-05-20 2012-10-23 Manyworlds, Inc. Adaptive commerce systems and methods
US8380579B2 (en) 2004-05-20 2013-02-19 Manyworlds, Inc. Contextual commerce systems and methods
US7904341B2 (en) 2004-05-20 2011-03-08 Manyworlds, Inc. Adaptive commerce systems and methods
US20070073553A1 (en) * 2004-05-20 2007-03-29 Manyworlds, Inc. Adaptive Commerce Systems and Methods
US20120096155A1 (en) * 2004-05-20 2012-04-19 Manyworlds, Inc. Subscription-based Temporally Sequenced Recommendations
US20130262667A1 (en) * 2004-06-18 2013-10-03 Fortinet, Inc. Systems and methods for categorizing network traffic content
US20060036728A1 (en) * 2004-06-18 2006-02-16 Fortinet, Inc. Systems and methods for categorizing network traffic content
US9237160B2 (en) 2004-06-18 2016-01-12 Fortinet, Inc. Systems and methods for categorizing network traffic content
US7979543B2 (en) * 2004-06-18 2011-07-12 Fortinet, Inc. Systems and methods for categorizing network traffic content
US20090234879A1 (en) * 2004-06-18 2009-09-17 Michael Xie Systems and methods for categorizing network traffic content
US20110231402A1 (en) * 2004-06-18 2011-09-22 Fortinet, Inc. Systems and methods for categorizing network traffic content
US20170187738A1 (en) * 2004-06-18 2017-06-29 Fortinet, Inc. Systems and methods for categorizing network traffic content
US9537871B2 (en) 2004-06-18 2017-01-03 Fortinet, Inc. Systems and methods for categorizing network traffic content
US8635336B2 (en) * 2004-06-18 2014-01-21 Fortinet, Inc. Systems and methods for categorizing network traffic content
US8782223B2 (en) * 2004-06-18 2014-07-15 Fortinet, Inc. Systems and methods for categorizing network traffic content
US7565445B2 (en) * 2004-06-18 2009-07-21 Fortinet, Inc. Systems and methods for categorizing network traffic content
US9223868B2 (en) * 2004-06-28 2015-12-29 Google Inc. Deriving and using interaction profiles
US20060015504A1 (en) * 2004-07-15 2006-01-19 Qingfeng Yu Method and system for site path evaluation using web session clustering
US8572233B2 (en) * 2004-07-15 2013-10-29 Hewlett-Packard Development Company, L.P. Method and system for site path evaluation using web session clustering
US7603345B2 (en) 2004-07-26 2009-10-13 Google Inc. Detecting spam documents in a phrase based information retrieval system
US9361331B2 (en) 2004-07-26 2016-06-07 Google Inc. Multiple index based information retrieval system
US8560550B2 (en) 2004-07-26 2013-10-15 Google, Inc. Multiple index based information retrieval system
US8108412B2 (en) 2004-07-26 2012-01-31 Google, Inc. Phrase-based detection of duplicate documents in an information retrieval system
US8078629B2 (en) 2004-07-26 2011-12-13 Google Inc. Detecting spam documents in a phrase based information retrieval system
US9569505B2 (en) 2004-07-26 2017-02-14 Google Inc. Phrase-based searching in an information retrieval system
US7702618B1 (en) 2004-07-26 2010-04-20 Google Inc. Information retrieval system for archiving multiple document versions
US7711679B2 (en) 2004-07-26 2010-05-04 Google Inc. Phrase-based detection of duplicate documents in an information retrieval system
US7536408B2 (en) 2004-07-26 2009-05-19 Google Inc. Phrase-based indexing in an information retrieval system
US9384224B2 (en) 2004-07-26 2016-07-05 Google Inc. Information retrieval system for archiving multiple document versions
US9037573B2 (en) 2004-07-26 2015-05-19 Google, Inc. Phase-based personalization of searches in an information retrieval system
US9817886B2 (en) 2004-07-26 2017-11-14 Google Llc Information retrieval system for archiving multiple document versions
US7580929B2 (en) * 2004-07-26 2009-08-25 Google Inc. Phrase-based personalization of searches in an information retrieval system
US20060106792A1 (en) * 2004-07-26 2006-05-18 Patterson Anna L Multiple index based information retrieval system
US7580921B2 (en) 2004-07-26 2009-08-25 Google Inc. Phrase identification in an information retrieval system
US7584175B2 (en) 2004-07-26 2009-09-01 Google Inc. Phrase-based generation of document descriptions
US9817825B2 (en) 2004-07-26 2017-11-14 Google Llc Multiple index based information retrieval system
US7599914B2 (en) 2004-07-26 2009-10-06 Google Inc. Phrase-based searching in an information retrieval system
US8489628B2 (en) 2004-07-26 2013-07-16 Google Inc. Phrase-based detection of duplicate documents in an information retrieval system
US7567959B2 (en) 2004-07-26 2009-07-28 Google Inc. Multiple index based information retrieval system
US7853606B1 (en) 2004-09-14 2010-12-14 Google, Inc. Alternate methods of displaying search results
US8635216B1 (en) * 2004-09-30 2014-01-21 Avaya Inc. Enhancing network information retrieval according to a user search profile
US20060074883A1 (en) * 2004-10-05 2006-04-06 Microsoft Corporation Systems, methods, and interfaces for providing personalized search and information access
US8019744B1 (en) * 2004-10-06 2011-09-13 Shopzilla, Inc. Search ranking estimation
US8473477B1 (en) * 2004-10-06 2013-06-25 Shopzilla, Inc. Search ranking estimation
US7539652B2 (en) 2004-11-04 2009-05-26 Manyworlds, Inc. Adaptive self-modifying and recombinant systems
US20060200433A1 (en) * 2004-11-04 2006-09-07 Manyworlds, Inc. Adaptive Self-Modifying and Recombinant Systems
US20080104039A1 (en) * 2004-11-24 2008-05-01 Linda Lowson System and method for resource management
US20080133573A1 (en) * 2004-12-24 2008-06-05 Michael Haft Relational Compressed Database Images (for Accelerated Querying of Databases)
US20110112986A1 (en) * 2005-01-18 2011-05-12 Manyworlds, Inc. Generative Investment Method and System
US20100169305A1 (en) * 2005-01-25 2010-07-01 Google Inc. Information retrieval system for archiving multiple document versions
US8612427B2 (en) * 2005-01-25 2013-12-17 Google, Inc. Information retrieval system for archiving multiple document versions
US9141589B2 (en) * 2005-03-09 2015-09-22 Google Inc. Method and an apparatus to provide a personalized page
US20110060736A1 (en) * 2005-03-29 2011-03-10 Google Inc. Query Revision Using Known Highly-Ranked Queries
US8375049B2 (en) 2005-03-29 2013-02-12 Google Inc. Query revision using known highly-ranked queries
US20060224554A1 (en) * 2005-03-29 2006-10-05 Bailey David R Query revision using known highly-ranked queries
US7870147B2 (en) 2005-03-29 2011-01-11 Google Inc. Query revision using known highly-ranked queries
US8140524B1 (en) 2005-03-30 2012-03-20 Google Inc. Estimating confidence for query revision models
US9069841B1 (en) 2005-03-30 2015-06-30 Google Inc. Estimating confidence for query revision models
US7779073B2 (en) * 2005-03-31 2010-08-17 British Telecommunications Plc Computer network
US20090037566A1 (en) * 2005-03-31 2009-02-05 British Telecommunications Public Limited Company Computer Network
US20070094157A1 (en) * 2005-04-08 2007-04-26 Manyworlds, Inc. Adaptive Digital Asset and Media Management
US20070203589A1 (en) * 2005-04-08 2007-08-30 Manyworlds, Inc. Adaptive Recombinant Process Methods
US9003441B1 (en) * 2005-04-13 2015-04-07 Google Inc. Probabilistic inference of demographic information from user selection of content
US20110093341A1 (en) * 2005-05-16 2011-04-21 Manyworlds, Inc. Explanatory Advertising Systems and Methods
US20060271846A1 (en) * 2005-05-24 2006-11-30 Microsoft Corporation Systems and methods that facilitate improved display of electronic documents
US7661065B2 (en) * 2005-05-24 2010-02-09 Microsoft Corporation Systems and methods that facilitate improved display of electronic documents
US20070011203A1 (en) * 2005-05-31 2007-01-11 Tomohiro Tsunoda Information processing device, method of information processing, and program
US8103659B1 (en) * 2005-06-06 2012-01-24 A9.Com, Inc. Perspective-based item navigation
US20060288017A1 (en) * 2005-06-17 2006-12-21 Goetz Oliver C User-specific product documentation
US20070005646A1 (en) * 2005-06-30 2007-01-04 Microsoft Corporation Analysis of topic dynamics of web search
WO2007005465A3 (en) * 2005-06-30 2008-06-26 Microsoft Corp Analysis of topic dynamics of web search
WO2007005465A2 (en) * 2005-06-30 2007-01-11 Microsoft Corporation Analysis of topic dynamics of web search
US9183752B2 (en) * 2005-07-14 2015-11-10 Red Hat, Inc. Tutorial generator with automatic capture of screenshots
US20070015118A1 (en) * 2005-07-14 2007-01-18 Red Hat, Inc. Tutorial generator with automatic capture of screenshots
US20070022082A1 (en) * 2005-07-20 2007-01-25 International Business Machines Corporation Search engine coverage
US20070027864A1 (en) * 2005-07-29 2007-02-01 Collins Robert J System and method for determining semantically related terms
US20070027865A1 (en) * 2005-07-29 2007-02-01 Kevin Bartz System and method for determining semantically related term
US20090198684A1 (en) * 2005-07-29 2009-08-06 Yahoo! Inc. System and Method for Determining Semantically Related Terms
US7548929B2 (en) 2005-07-29 2009-06-16 Yahoo! Inc. System and method for determining semantically related terms
US20070038634A1 (en) * 2005-08-09 2007-02-15 Glover Eric J Method for targeting World Wide Web content and advertising to a user
US9105028B2 (en) 2005-08-10 2015-08-11 Compete, Inc. Monitoring clickstream behavior of viewers of online advertisements and search results
US20110022564A1 (en) * 2005-11-02 2011-01-27 Manyworlds, Inc. Adaptive Knowledge Lifecycle Management Methods
US7831535B2 (en) 2005-11-02 2010-11-09 Manyworlds, Inc. Adaptive knowledge lifecycle management aligned with assessed financial value based on organizing structure of knowledge assets
US20070112712A1 (en) * 2005-11-02 2007-05-17 Manyworlds, Inc. Adaptive Knowledge Lifecycle Management
US8577880B1 (en) 2005-11-17 2013-11-05 Amazon Technologies, Inc. Recommendations based on item tagging activities of users
US8122020B1 (en) 2005-11-17 2012-02-21 Amazon Technologies, Inc. Recommendations based on item tagging activities of users
US8341158B2 (en) * 2005-11-21 2012-12-25 Sony Corporation User's preference prediction from collective rating data
US20070118546A1 (en) * 2005-11-21 2007-05-24 Sony Corporation User's preference prediction from collective rating data
US9262446B1 (en) 2005-12-29 2016-02-16 Google Inc. Dynamically ranking entries in a personal data book
US20070168342A1 (en) * 2005-12-29 2007-07-19 Singerman Brian A Recommended alerts
US8799302B2 (en) * 2005-12-29 2014-08-05 Google Inc. Recommended alerts
US9323846B2 (en) 2005-12-30 2016-04-26 Google Inc. Method, system, and graphical user interface for alerting a computer user to new results for a prior search
US8694491B2 (en) 2005-12-30 2014-04-08 Google Inc. Method, system, and graphical user interface for alerting a computer user to new results for a prior search
US7925649B2 (en) 2005-12-30 2011-04-12 Google Inc. Method, system, and graphical user interface for alerting a computer user to new results for a prior search
US20070162424A1 (en) * 2005-12-30 2007-07-12 Glen Jeh Method, system, and graphical user interface for alerting a computer user to new results for a prior search
US20110137849A1 (en) * 2006-01-10 2011-06-09 Manyworlds, Inc. Adaptive Experimentation Method and System
US9159027B2 (en) 2006-01-10 2015-10-13 Manyworlds, Inc. Adaptive experimentation method and system
US20080059395A1 (en) * 2006-01-10 2008-03-06 Manyworlds, Inc. Adaptive Online Experimentation
US9092503B2 (en) 2006-03-06 2015-07-28 Veveo, Inc. Methods and systems for selecting and presenting content based on dynamically identifying microgenres associated with the content
US9128987B2 (en) 2006-03-06 2015-09-08 Veveo, Inc. Methods and systems for selecting and presenting content based on a comparison of preference signatures from multiple users
US9075861B2 (en) 2006-03-06 2015-07-07 Veveo, Inc. Methods and systems for segmenting relative user preferences into fine-grain and coarse-grain collections
US8863221B2 (en) 2006-03-07 2014-10-14 Samsung Electronics Co., Ltd. Method and system for integrating content and services among multiple networks
US8200688B2 (en) * 2006-03-07 2012-06-12 Samsung Electronics Co., Ltd. Method and system for facilitating information searching on electronic devices
US9749693B2 (en) 2006-03-24 2017-08-29 Rovi Guides, Inc. Interactive media guidance application with intelligent navigation and display features
WO2007120895A3 (en) * 2006-04-14 2008-12-18 At & T Corp Automatic learning for mapping spoken/text descriptions of products onto available products
US20100088160A1 (en) * 2006-04-14 2010-04-08 At&T Corp. Automatic Learning for Mapping Spoken/Text Descriptions of Products onto Available Products
US7630917B2 (en) * 2006-04-14 2009-12-08 At&T Intellectual Property Ii, L.P. Automatic learning for mapping spoken/text descriptions of products onto available products
US7937301B2 (en) * 2006-04-14 2011-05-03 At&T Intellectual Property Ii, L.P. Automatic learning for mapping spoken/text descriptions of products onto available products
US20070250404A1 (en) * 2006-04-14 2007-10-25 At&T Corp. Automatic Learning For Mapping Spoken/Text Descriptions Of Products Onto Available Products
US20070255701A1 (en) * 2006-04-28 2007-11-01 Halla Jason M System and method for analyzing internet content and correlating to events
US20120078710A1 (en) * 2006-05-02 2012-03-29 Mark Cramer Dynamic search engine results employing user behavior
US20100106703A1 (en) * 2006-05-02 2010-04-29 Mark Cramer Dynamic search engine results employing user behavior
US8442973B2 (en) * 2006-05-02 2013-05-14 Surf Canyon, Inc. Real time implicit user modeling for personalized search
US8095582B2 (en) * 2006-05-02 2012-01-10 Surf Canyon Incorporated Dynamic search engine results employing user behavior
US20080114751A1 (en) * 2006-05-02 2008-05-15 Surf Canyon Incorporated Real time implicit user modeling for personalized search
WO2007133334A1 (en) * 2006-05-11 2007-11-22 Yahoo! Inc. System and method for determining semantically related terms
US20100030713A1 (en) * 2006-05-24 2010-02-04 Icom Limited Content engine
US7827125B1 (en) * 2006-06-01 2010-11-02 Trovix, Inc. Learning based on feedback for contextual personalized information retrieval
US8463810B1 (en) 2006-06-01 2013-06-11 Monster Worldwide, Inc. Scoring concepts for contextual personalized information retrieval
US8001064B1 (en) * 2006-06-01 2011-08-16 Monster Worldwide, Inc. Learning based on feedback for contextual personalized information retrieval
US7870117B1 (en) 2006-06-01 2011-01-11 Monster Worldwide, Inc. Constructing a search query to execute a contextual personalized search of a knowledge base
US8024329B1 (en) 2006-06-01 2011-09-20 Monster Worldwide, Inc. Using inverted indexes for contextual personalized information retrieval
US9443022B2 (en) 2006-06-05 2016-09-13 Google Inc. Method, system, and graphical user interface for providing personalized recommendations of popular search queries
US20070288431A1 (en) * 2006-06-09 2007-12-13 Ebay Inc. System and method for application programming interfaces for keyword extraction and contextual advertisement generation
US8001105B2 (en) 2006-06-09 2011-08-16 Ebay Inc. System and method for keyword extraction and contextual advertisement generation
US8200662B2 (en) 2006-06-09 2012-06-12 Ebay Inc. System and method for application programming interfaces for keyword extraction and contextual advertisement generation
US20070288454A1 (en) * 2006-06-09 2007-12-13 Ebay Inc. System and method for keyword extraction and contextual advertisement generation
US7831586B2 (en) 2006-06-09 2010-11-09 Ebay Inc. System and method for application programming interfaces for keyword extraction and contextual advertisement generation
US9495452B2 (en) 2006-06-22 2016-11-15 Yahoo! Inc. User-sensitive PageRank
US20080010258A1 (en) * 2006-06-22 2008-01-10 Infosys Technologies Ltd. Collaborative filtering-based recommendations
US20080010281A1 (en) * 2006-06-22 2008-01-10 Yahoo! Inc. User-sensitive pagerank
US7624104B2 (en) * 2006-06-22 2009-11-24 Yahoo! Inc. User-sensitive pagerank
US20100023513A1 (en) * 2006-06-22 2010-01-28 Yahoo! Inc. User-sensitive pagerank
US8468155B2 (en) * 2006-06-22 2013-06-18 Infosys Limited Collaborative filtering-based recommendations
US20080040218A1 (en) * 2006-07-05 2008-02-14 Van Dijk Bob System and method for category-based contextual advertisement generation and management
US20090177546A1 (en) * 2006-07-05 2009-07-09 Dijk Bob Van System and method for category-based contextual advertisement generation and management
US20090177531A1 (en) * 2006-07-05 2009-07-09 Dijk Bob Van System and method for category-based contextual advertisement generation and management
US20100094859A1 (en) * 2006-08-09 2010-04-15 Gupta Puneet K Folksonomy-Enhanced Enterprise-Centric Collaboration and Knowledge Management System
US8135725B2 (en) * 2006-08-11 2012-03-13 Yahoo! Inc. System and method for providing tag-based relevance recommendations of bookmarks in a bookmark and tag database
US20080040313A1 (en) * 2006-08-11 2008-02-14 Yahoo! Inc. System and method for providing tag-based relevance recommendations of bookmarks in a bookmark and tag database
US9418122B2 (en) * 2006-08-25 2016-08-16 Surf Canyon Incorporated Adaptive user interface for real-time search relevance feedback
US8924378B2 (en) * 2006-08-25 2014-12-30 Surf Canyon Incorporated Adaptive user interface for real-time search relevance feedback
US20150081691A1 (en) * 2006-08-25 2015-03-19 Surf Canyon Incorporated Adaptive user interface for real-time search relevance feedback
US8458120B2 (en) 2006-08-28 2013-06-04 Manyworlds, Inc. Search-based people matching system and method
US8515901B2 (en) 2006-08-28 2013-08-20 Manyworlds, Inc. Explanatory people matching system and method
US20100217722A1 (en) * 2006-08-28 2010-08-26 Manyworlds, Inc. Mutual Interest Inferencing System and Method
US8515900B2 (en) 2006-08-28 2013-08-20 Manyworlds, Inc. Environment-responsive people matching system and method
US8060462B2 (en) 2006-08-28 2011-11-15 Manyworlds, Inc. Mutual interest inferencing system and method
US8478716B2 (en) 2006-08-28 2013-07-02 Manyworlds, Inc. Proximity-based people matching system and method
US8458119B2 (en) 2006-08-28 2013-06-04 Manyworlds, Inc. People matching in subscription-based communities
WO2008032037A1 (en) * 2006-09-11 2008-03-20 Stevens Richard J Method and system for filtering and searching data using word frequencies
US20080097821A1 (en) * 2006-10-24 2008-04-24 Microsoft Corporation Recommendations utilizing meta-data based pair-wise lift predictions
US20080120072A1 (en) * 2006-11-16 2008-05-22 Yahoo! Inc. System and method for determining semantically related terms based on sequences of search queries
US7814086B2 (en) 2006-11-16 2010-10-12 Yahoo! Inc. System and method for determining semantically related terms based on sequences of search queries
US20080120292A1 (en) * 2006-11-20 2008-05-22 Neelakantan Sundaresan Search clustering
US8131722B2 (en) * 2006-11-20 2012-03-06 Ebay Inc. Search clustering
US8589398B2 (en) 2006-11-20 2013-11-19 Ebay Inc. Search clustering
US8935269B2 (en) 2006-12-04 2015-01-13 Samsung Electronics Co., Ltd. Method and apparatus for contextual search and query refinement on consumer electronics devices
EP2020796A1 (en) * 2006-12-15 2009-02-04 Huawei Technologies Co Ltd A method for accessing user data and a user profile management server
US20090254561A1 (en) * 2006-12-15 2009-10-08 Shao Gang Method for Accessing User Data and Profile Management Server
EP2020796A4 (en) * 2006-12-15 2009-05-27 Huawei Tech Co Ltd A method for accessing user data and a user profile management server
US20100122257A1 (en) * 2007-01-26 2010-05-13 Kyocera Corporation Electronic device and electronic device control method
US8631413B2 (en) * 2007-01-26 2014-01-14 Kyocera Corporation Determining the termination priority of applications based on capability of applications to retain operation state information
US20080183681A1 (en) * 2007-01-29 2008-07-31 Samsung Electronics Co., Ltd. Method and system for facilitating information searching on electronic devices
US8782056B2 (en) 2007-01-29 2014-07-15 Samsung Electronics Co., Ltd. Method and system for facilitating information searching on electronic devices
US9177048B1 (en) 2007-02-16 2015-11-03 Google Inc. Collaborative filtering
US8407219B1 (en) 2007-02-16 2013-03-26 Google Inc. Collaborative filtering
US8407226B1 (en) 2007-02-16 2013-03-26 Google Inc. Collaborative filtering
US8135718B1 (en) 2007-02-16 2012-03-13 Google Inc. Collaborative filtering
US8065254B1 (en) 2007-02-19 2011-11-22 Google Inc. Presenting a diversity of recommendations
US8374985B1 (en) 2007-02-19 2013-02-12 Google Inc. Presenting a diversity of recommendations
US8115869B2 (en) 2007-02-28 2012-02-14 Samsung Electronics Co., Ltd. Method and system for extracting relevant information from content metadata
US20080221987A1 (en) * 2007-03-07 2008-09-11 Ebay Inc. System and method for contextual advertisement and merchandizing based on an automatically generated user demographic profile
US20080225870A1 (en) * 2007-03-15 2008-09-18 Sundstrom Robert J Methods, systems, and computer program products for providing predicted likelihood of communication between users
US9686367B2 (en) * 2007-03-15 2017-06-20 Scenera Technologies, Llc Methods, systems, and computer program products for providing predicted likelihood of communication between users
US8510453B2 (en) 2007-03-21 2013-08-13 Samsung Electronics Co., Ltd. Framework for correlating content on a local network with information on an external network
US20100049663A1 (en) * 2007-03-30 2010-02-25 Kane Jr Francis J Service for providing item recommendations
US7925655B1 (en) 2007-03-30 2011-04-12 Google Inc. Query scheduling using hierarchical tiers of index servers
US8682901B1 (en) 2007-03-30 2014-03-25 Google Inc. Index server architecture using tiered and sharded phrase posting lists
US8504441B2 (en) 2007-03-30 2013-08-06 Amazon Technologies, Inc. Services for providing item association data
US7945485B2 (en) 2007-03-30 2011-05-17 Amazon Technologies, Inc. Service for providing item recommendations
US7702614B1 (en) 2007-03-30 2010-04-20 Google Inc. Index updating using segment swapping
US7693813B1 (en) 2007-03-30 2010-04-06 Google Inc. Index server architecture using tiered and sharded phrase posting lists
US7881984B2 (en) 2007-03-30 2011-02-01 Amazon Technologies, Inc. Service for providing item recommendations
US8402033B1 (en) 2007-03-30 2013-03-19 Google Inc. Phrase extraction using subphrase scoring
US20110119150A1 (en) * 2007-03-30 2011-05-19 Kane Jr Francis J Outsourced generation of behavior-based content
US7974888B2 (en) 2007-03-30 2011-07-05 Amazon Technologies, Inc. Services for providing item association data
US9355169B1 (en) 2007-03-30 2016-05-31 Google Inc. Phrase extraction using subphrase scoring
US9652483B1 (en) 2007-03-30 2017-05-16 Google Inc. Index server architecture using tiered and sharded phrase posting lists
US8275674B2 (en) 2007-03-30 2012-09-25 Amazon Technologies, Inc. Services for providing item association data
US20080243826A1 (en) * 2007-03-30 2008-10-02 Yahoo! Inc. System and method for determining semantically related terms
US8943067B1 (en) 2007-03-30 2015-01-27 Google Inc. Index server architecture using tiered and sharded phrase posting lists
US8073747B2 (en) 2007-03-30 2011-12-06 Amazon Technologies, Inc. Outsourced generation of behavior-based content
US20080243632A1 (en) * 2007-03-30 2008-10-02 Kane Francis J Service for providing item recommendations
US20080243480A1 (en) * 2007-03-30 2008-10-02 Yahoo! Inc. System and method for determining semantically related terms
US8166021B1 (en) 2007-03-30 2012-04-24 Google Inc. Query phrasification
US8166045B1 (en) 2007-03-30 2012-04-24 Google Inc. Phrase extraction using subphrase scoring
US8086594B1 (en) 2007-03-30 2011-12-27 Google Inc. Bifurcated document relevance scoring
US9223877B1 (en) 2007-03-30 2015-12-29 Google Inc. Index server architecture using tiered and sharded phrase posting lists
US8090723B2 (en) 2007-03-30 2012-01-03 Google Inc. Index server architecture using tiered and sharded phrase posting lists
WO2008121737A1 (en) * 2007-03-30 2008-10-09 Amazon Technologies, Inc. Service for providing item recommendations
US8600975B1 (en) 2007-03-30 2013-12-03 Google Inc. Query phrasification
US20080243631A1 (en) * 2007-03-30 2008-10-02 Kane Francis J Services for providing item association data
US20080250323A1 (en) * 2007-04-04 2008-10-09 Huff Gerald B Method and apparatus for recommending an application-feature to a user
US8209724B2 (en) 2007-04-25 2012-06-26 Samsung Electronics Co., Ltd. Method and system for providing access to information of potential interest to a user
US9286385B2 (en) 2007-04-25 2016-03-15 Samsung Electronics Co., Ltd. Method and system for providing access to information of potential interest to a user
US20100169249A1 (en) * 2007-04-30 2010-07-01 Yahoo Inc. System and Method for Determining Semantically Related Terms Using an Active Learning Framework
US7707127B2 (en) 2007-04-30 2010-04-27 Yahoo! Inc. Method and apparatus using a classifier to determine semantically relevant terms
US8275722B2 (en) 2007-04-30 2012-09-25 Yahoo! Inc. System and method for determining semantically related terms using an active learning framework
US20080270333A1 (en) * 2007-04-30 2008-10-30 Yahoo! Inc. System and method for determining semantically related terms using an active learning framework
US8843467B2 (en) 2007-05-15 2014-09-23 Samsung Electronics Co., Ltd. Method and system for providing relevant information to a user of a device in a local network
US20080289039A1 (en) * 2007-05-18 2008-11-20 Sap Ag Method and system for protecting a message from an xml attack when being exchanged in a distributed and decentralized network system
US8316443B2 (en) * 2007-05-18 2012-11-20 Sap Ag Method and system for protecting a message from an XML attack when being exchanged in a distributed and decentralized network system
US20080307053A1 (en) * 2007-06-08 2008-12-11 Mitnick Craig R System and Method for Permitting Geographically-Pertinent Information to be Ranked by Users According to Users' Geographic Proximity to Information and to Each Other for Affecting the Ranking of Such Information
US7739261B2 (en) 2007-06-14 2010-06-15 Microsoft Corporation Identification of topics for online discussions based on language patterns
US20080313180A1 (en) * 2007-06-14 2008-12-18 Microsoft Corporation Identification of topics for online discussions based on language patterns
US20090006398A1 (en) * 2007-06-29 2009-01-01 Shing Yan Lam Recommendation system with multiple integrated recommenders
US8260787B2 (en) 2007-06-29 2012-09-04 Amazon Technologies, Inc. Recommendation system with multiple integrated recommenders
US8751507B2 (en) * 2007-06-29 2014-06-10 Amazon Technologies, Inc. Recommendation system with multiple integrated recommenders
US9392074B2 (en) 2007-07-07 2016-07-12 Qualcomm Incorporated User profile generation architecture for mobile content-message targeting
US9596317B2 (en) 2007-07-07 2017-03-14 Qualcomm Incorporated Method and system for delivery of targeted information based on a user profile in a mobile communication device
US9497286B2 (en) 2007-07-07 2016-11-15 Qualcomm Incorporated Method and system for providing targeted information based on a user profile in a mobile environment
US9485322B2 (en) 2007-07-07 2016-11-01 Qualcomm Incorporated Method and system for providing targeted information using profile attributes with variable confidence levels in a mobile environment
US9398113B2 (en) 2007-07-07 2016-07-19 Qualcomm Incorporated Methods and systems for providing targeted information using identity masking in a wireless communications device
US20090018904A1 (en) * 2007-07-09 2009-01-15 Ebay Inc. System and method for contextual advertising and merchandizing based on user configurable preferences
US20090037399A1 (en) * 2007-07-31 2009-02-05 Yahoo! Inc. System and Method for Determining Semantically Related Terms
US20100131333A1 (en) * 2007-08-06 2010-05-27 Keith Ball Method and system for product services analysis and optimization
EP2176730A1 (en) * 2007-08-08 2010-04-21 Baynote, Inc. Method and apparatus for context-based content recommendation
EP2176730A4 (en) * 2007-08-08 2011-04-20 Baynote Inc Method and apparatus for context-based content recommendation
US20090043760A1 (en) * 2007-08-10 2009-02-12 Kabushiki Kaisha Toshiba Program searching apparatus and program searching method
EP2023634A1 (en) * 2007-08-10 2009-02-11 Kabushiki Kaisha Toshiba Program searching apparatus and program searching method
US9471616B2 (en) 2007-08-28 2016-10-18 International Business Machines Corporation Managing user ratings in a web services environment
US20090063408A1 (en) * 2007-08-28 2009-03-05 International Business Machines Corporation Managing user ratings in a web services environment
US8635222B2 (en) 2007-08-28 2014-01-21 International Business Machines Corporation Managing user ratings in a web services environment
US9165075B2 (en) 2007-08-28 2015-10-20 International Business Machines Corporation Managing user ratings in a web services environment
US8631027B2 (en) 2007-09-07 2014-01-14 Google Inc. Integrated external related phrase information into a phrase-based indexing information retrieval system
US8117223B2 (en) 2007-09-07 2012-02-14 Google Inc. Integrating external related phrase information into a phrase-based indexing information retrieval system
US20090144654A1 (en) * 2007-10-03 2009-06-04 Robert Brouwer Methods and apparatus for facilitating content consumption
US8671104B2 (en) 2007-10-12 2014-03-11 Palo Alto Research Center Incorporated System and method for providing orientation into digital information
US20090099996A1 (en) * 2007-10-12 2009-04-16 Palo Alto Research Center Incorporated System And Method For Performing Discovery Of Digital Information In A Subject Area
US8930388B2 (en) 2007-10-12 2015-01-06 Palo Alto Research Center Incorporated System and method for providing orientation into subject areas of digital information for augmented communities
US8073682B2 (en) 2007-10-12 2011-12-06 Palo Alto Research Center Incorporated System and method for prospecting digital information
US8165985B2 (en) 2007-10-12 2012-04-24 Palo Alto Research Center Incorporated System and method for performing discovery of digital information in a subject area
US20090100043A1 (en) * 2007-10-12 2009-04-16 Palo Alto Research Center Incorporated System And Method For Providing Orientation Into Digital Information
US20090099839A1 (en) * 2007-10-12 2009-04-16 Palo Alto Research Center Incorporated System And Method For Prospecting Digital Information
US8706678B2 (en) 2007-10-12 2014-04-22 Palo Alto Research Center Incorporated System and method for facilitating evergreen discovery of digital information
US8190424B2 (en) 2007-10-12 2012-05-29 Palo Alto Research Center Incorporated Computer-implemented system and method for prospecting digital information through online social communities
US8176068B2 (en) 2007-10-31 2012-05-08 Samsung Electronics Co., Ltd. Method and system for suggesting search queries on electronic devices
US20090125934A1 (en) * 2007-11-11 2009-05-14 Microsoft Corporation User rating mechanism for media content
US9203912B2 (en) 2007-11-14 2015-12-01 Qualcomm Incorporated Method and system for message value calculation in a mobile environment
US9705998B2 (en) 2007-11-14 2017-07-11 Qualcomm Incorporated Method and system using keyword vectors and associated metrics for learning and prediction of user correlation of targeted content messages in a mobile environment
US9203911B2 (en) 2007-11-14 2015-12-01 Qualcomm Incorporated Method and system for using a cache miss state match indicator to determine user suitability of targeted content messages in a mobile environment
US20090132526A1 (en) * 2007-11-19 2009-05-21 Jong-Hun Park Content recommendation apparatus and method using tag cloud
US8209337B2 (en) * 2007-11-19 2012-06-26 Core Logic, Inc. Content recommendation apparatus and method using tag cloud
US9391789B2 (en) 2007-12-14 2016-07-12 Qualcomm Incorporated Method and system for multi-level distribution information cache management in a mobile environment
US9015147B2 (en) 2007-12-20 2015-04-21 Porto Technology, Llc System and method for generating dynamically filtered content results, including for audio and/or video channels
US9311364B2 (en) 2007-12-20 2016-04-12 Porto Technology, Llc System and method for generating dynamically filtered content results, including for audio and/or video channels
US9552428B2 (en) 2007-12-21 2017-01-24 Lemi Technology, Llc System for generating media recommendations in a distributed environment based on seed information
US8316015B2 (en) 2007-12-21 2012-11-20 Lemi Technology, Llc Tunersphere
US8983937B2 (en) 2007-12-21 2015-03-17 Lemi Technology, Llc Tunersphere
US8117193B2 (en) 2007-12-21 2012-02-14 Lemi Technology, Llc Tunersphere
US9275138B2 (en) 2007-12-21 2016-03-01 Lemi Technology, Llc System for generating media recommendations in a distributed environment based on seed information
US8577874B2 (en) 2007-12-21 2013-11-05 Lemi Technology, Llc Tunersphere
US8874554B2 (en) 2007-12-21 2014-10-28 Lemi Technology, Llc Turnersphere
US20090172551A1 (en) * 2007-12-28 2009-07-02 Kane Francis J Behavior-based selection of items to present on affiliate sites
US8271878B2 (en) 2007-12-28 2012-09-18 Amazon Technologies, Inc. Behavior-based selection of items to present on affiliate sites
US20090171968A1 (en) * 2007-12-28 2009-07-02 Kane Francis J Widget-assisted content personalization based on user behaviors tracked across multiple web sites
US7984056B1 (en) 2007-12-28 2011-07-19 Amazon Technologies, Inc. System for facilitating discovery and management of feeds
US20090172021A1 (en) * 2007-12-28 2009-07-02 Kane Francis J Recommendations based on actions performed on multiple remote servers
US20090171755A1 (en) * 2007-12-28 2009-07-02 Kane Francis J Behavior-based generation of site-to-site referrals
US8589418B1 (en) 2007-12-28 2013-11-19 Amazon Technologies, Inc. System for facilitating discovery and management of feeds
US20090171754A1 (en) * 2007-12-28 2009-07-02 Kane Francis J Widget-assisted detection and exposure of cross-site behavioral associations
US20110246174A1 (en) * 2008-01-17 2011-10-06 Geacom, Inc. Method and system for situational language interpretation
US9418062B2 (en) * 2008-01-17 2016-08-16 Geacom, Inc. Method and system for situational language interpretation
US8782124B2 (en) * 2008-02-28 2014-07-15 Red Hat, Inc. Enhancing browser history using syndicated data feeds
US20090222735A1 (en) * 2008-02-28 2009-09-03 Clark Bryan W Systems and methods for enhancing browser history using syndicated data feeds
US20100005424A1 (en) * 2008-03-05 2010-01-07 Neelakantan Sundaresan Virtual world system supporting a consumer experience
US9081853B2 (en) 2008-04-03 2015-07-14 Graham Holdings Company Information display system based on user profile data with assisted and explicit profile modification
US20090254838A1 (en) * 2008-04-03 2009-10-08 Icurrent, Inc. Information display system based on user profile data with assisted and explicit profile modification
US20110035381A1 (en) * 2008-04-23 2011-02-10 Simon Giles Thompson Method
US8825650B2 (en) * 2008-04-23 2014-09-02 British Telecommunications Public Limited Company Method of classifying and sorting online content
US8402387B1 (en) 2008-04-24 2013-03-19 Google Inc. Advertising based on user models
US8214364B2 (en) * 2008-05-21 2012-07-03 International Business Machines Corporation Modeling user access to computer resources
US20090292743A1 (en) * 2008-05-21 2009-11-26 Bigus Joseph P Modeling user access to computer resources
US8359237B2 (en) * 2008-05-23 2013-01-22 Ebay Inc. System and method for context and community based customization for a user experience
US20090292584A1 (en) * 2008-05-23 2009-11-26 Ebay Inc. System and method for context and community based customization for a user experience
US8543570B1 (en) * 2008-06-10 2013-09-24 Surf Canyon Incorporated Adaptive user interface for real-time search relevance feedback
US8521731B2 (en) 2008-07-09 2013-08-27 Yahoo! Inc. Systems and methods for query expansion in sponsored search
US8533067B1 (en) 2008-08-12 2013-09-10 Amazon Technologies, Inc. System for obtaining recommendations from multiple recommenders
US7991650B2 (en) 2008-08-12 2011-08-02 Amazon Technologies, Inc. System for obtaining recommendations from multiple recommenders
US8249948B1 (en) 2008-08-12 2012-08-21 Amazon Technologies, Inc. System for obtaining recommendations from multiple recommenders
US7991757B2 (en) 2008-08-12 2011-08-02 Amazon Technologies, Inc. System for obtaining recommendations from multiple recommenders
US20100042608A1 (en) * 2008-08-12 2010-02-18 Kane Jr Francis J System for obtaining recommendations from multiple recommenders
US20100042460A1 (en) * 2008-08-12 2010-02-18 Kane Jr Francis J System for obtaining recommendations from multiple recommenders
US8090794B1 (en) * 2008-08-25 2012-01-03 Intuit Inc. Technique for customizing displayed content
US20100058195A1 (en) * 2008-08-28 2010-03-04 Palo Alto Research Center Incorporated System And Method For Interfacing A Web Browser Widget With Social Indexing
EP2159715A3 (en) * 2008-08-28 2010-04-21 Palo Alto Research Center Incorporated System and method for providing a topic-directed search
US20100057577A1 (en) * 2008-08-28 2010-03-04 Palo Alto Research Center Incorporated System And Method For Providing Topic-Guided Broadening Of Advertising Targets In Social Indexing
US20100057536A1 (en) * 2008-08-28 2010-03-04 Palo Alto Research Center Incorporated System And Method For Providing Community-Based Advertising Term Disambiguation
US20100057716A1 (en) * 2008-08-28 2010-03-04 Stefik Mark J System And Method For Providing A Topic-Directed Search
US8010545B2 (en) 2008-08-28 2011-08-30 Palo Alto Research Center Incorporated System and method for providing a topic-directed search
US8209616B2 (en) 2008-08-28 2012-06-26 Palo Alto Research Center Incorporated System and method for interfacing a web browser widget with social indexing
US8938465B2 (en) 2008-09-10 2015-01-20 Samsung Electronics Co., Ltd. Method and system for utilizing packaged content sources to identify and provide information based on contextual information
EP2328344A1 (en) * 2008-09-23 2011-06-01 Huawei Device Co., Ltd. Method, apparatus and system for playing programs
US8464294B2 (en) 2008-09-23 2013-06-11 Huawei Device Co., Ltd. Method, terminal and system for playing programs
US20110173666A1 (en) * 2008-09-23 2011-07-14 Huawei Display Co., Ltd. Method, terminal and system for playing programs
EP2328344A4 (en) * 2008-09-23 2012-02-29 Huawei Device Co Ltd Method, apparatus and system for playing programs
US8904345B2 (en) 2008-09-30 2014-12-02 Ebay Inc. System and method for orchestration of customization for a user experience
US20100083217A1 (en) * 2008-09-30 2010-04-01 Dalal Vipul C System and method for orchestration of customization for a user expereince
US9753902B2 (en) 2008-09-30 2017-09-05 Ebay Inc. System and method for orchestration of customization for a user experience
US20100114654A1 (en) * 2008-10-31 2010-05-06 Hewlett-Packard Development Company, L.P. Learning user purchase intent from user-centric data
US20100125540A1 (en) * 2008-11-14 2010-05-20 Palo Alto Research Center Incorporated System And Method For Providing Robust Topic Identification In Social Indexes
US20100125597A1 (en) * 2008-11-14 2010-05-20 Yahoo! Inc. System and method for determining search terms for use in sponsored searches
US8549016B2 (en) 2008-11-14 2013-10-01 Palo Alto Research Center Incorporated System and method for providing robust topic identification in social indexes
US8494899B2 (en) 2008-12-02 2013-07-23 Lemi Technology, Llc Dynamic talk radio program scheduling
US8805861B2 (en) 2008-12-09 2014-08-12 Google Inc. Methods and systems to train models to extract and integrate information from data sources
US20100145902A1 (en) * 2008-12-09 2010-06-10 Ita Software, Inc. Methods and systems to train models to extract and integrate information from data sources
US9317563B1 (en) * 2009-01-09 2016-04-19 Google Inc. Preferred sites
US8595228B1 (en) 2009-01-09 2013-11-26 Google Inc. Preferred sites
US20100180200A1 (en) * 2009-01-12 2010-07-15 Thierry Donneau-Golencer Electronic assistant
US9195640B1 (en) 2009-01-12 2015-11-24 Sri International Method and system for finding content having a desired similarity
US20100185473A1 (en) * 2009-01-20 2010-07-22 Microsoft Corporation Document vault and application platform
US8620778B2 (en) * 2009-01-20 2013-12-31 Microsoft Corporation Document vault and application platform
US20140101011A1 (en) * 2009-01-20 2014-04-10 Microsoft Corporation Document Vault and Application Platform
US20100191773A1 (en) * 2009-01-27 2010-07-29 Palo Alto Research Center Incorporated System And Method For Providing Default Hierarchical Training For Social Indexing
US20100191741A1 (en) * 2009-01-27 2010-07-29 Palo Alto Research Center Incorporated System And Method For Using Banded Topic Relevance And Time For Article Prioritization
US8239397B2 (en) 2009-01-27 2012-08-07 Palo Alto Research Center Incorporated System and method for managing user attention by detecting hot and cold topics in social indexes
US8356044B2 (en) 2009-01-27 2013-01-15 Palo Alto Research Center Incorporated System and method for providing default hierarchical training for social indexing
US20100191742A1 (en) * 2009-01-27 2010-07-29 Palo Alto Research Center Incorporated System And Method For Managing User Attention By Detecting Hot And Cold Topics In Social Indexes
US8452781B2 (en) 2009-01-27 2013-05-28 Palo Alto Research Center Incorporated System and method for using banded topic relevance and time for article prioritization
US20140236953A1 (en) * 2009-02-11 2014-08-21 Jeffrey A. Rapaport Methods using social topical adaptive networking system
US9600581B2 (en) * 2009-02-19 2017-03-21 Yahoo! Inc. Personalized recommendations on dynamic content
US20100211568A1 (en) * 2009-02-19 2010-08-19 Wei Chu Personalized recommendations on dynamic content
US9245033B2 (en) 2009-04-02 2016-01-26 Graham Holdings Company Channel sharing
US8463930B2 (en) 2009-04-29 2013-06-11 Lemi Technology, Llc Skip feature for a broadcast or multicast media station
US8977770B2 (en) 2009-04-29 2015-03-10 Lemi Technolgy, LLC Skip feature for a broadcast or multicast media station
US8806047B2 (en) 2009-04-29 2014-08-12 Lemi Technology, Llc Skip feature for a broadcast or multicast media station
US9432423B2 (en) 2009-04-29 2016-08-30 Lemi Technology, Llc Skip feature for a broadcast or multicast media station
US20100296728A1 (en) * 2009-05-22 2010-11-25 Shinya Ohtani Discrimination Apparatus, Method of Discrimination, and Computer Program
EP2271057A1 (en) * 2009-06-26 2011-01-05 Alcatel Lucent Device and method for exploiting exchanged data and actions of user communication equipment(s) to improve browsing dynamically, and associated network equipment
US9754288B2 (en) 2009-06-30 2017-09-05 Amazon Technologies, Inc. Recommendation of media content items based on geolocation and venue
US8856670B1 (en) * 2009-08-25 2014-10-07 Intuit Inc. Technique for customizing a user interface
US9785987B2 (en) 2010-04-22 2017-10-10 Microsoft Technology Licensing, Llc User interface for information presentation system
US9031944B2 (en) 2010-04-30 2015-05-12 Palo Alto Research Center Incorporated System and method for providing multi-core and multi-level topical organization in social indexes
US20110282861A1 (en) * 2010-05-11 2011-11-17 Microsoft Corporation Extracting higher-order knowledge from structured data
US20110295612A1 (en) * 2010-05-28 2011-12-01 Thierry Donneau-Golencer Method and apparatus for user modelization
US9785888B2 (en) * 2010-05-31 2017-10-10 Sony Corporation Information processing apparatus, information processing method, and program for prediction model generated based on evaluation information
EP2506522A1 (en) * 2010-08-04 2012-10-03 Tencent Technology (Shenzhen) Company Limited Method and device for pushing data
EP2506522A4 (en) * 2010-08-04 2014-11-19 Tencent Tech Shenzhen Co Ltd Method and device for pushing data
US9076146B2 (en) * 2010-10-15 2015-07-07 At&T Intellectual Property I, L.P. Personal customer care agent
US20120095861A1 (en) * 2010-10-15 2012-04-19 At&T Intellectual Property I, L.P. Personal customer care agent
US20130268317A1 (en) * 2010-12-07 2013-10-10 Digital Foodie Oy Arrangement for facilitating shopping and related method
US9003015B2 (en) 2010-12-21 2015-04-07 Sitecore A/S Method and a system for managing a website using profile key patterns
US9736524B2 (en) 2011-01-06 2017-08-15 Veveo, Inc. Methods of and systems for content search based on environment sampling
US9805127B2 (en) 2011-01-20 2017-10-31 Linkedin Corporation Methods and systems for utilizing activity data with clustered events
US9172762B2 (en) * 2011-01-20 2015-10-27 Linkedin Corporation Methods and systems for recommending a context based on content interaction
US20120191776A1 (en) * 2011-01-20 2012-07-26 Linkedin Corporation Methods and systems for recommending a context based on content interaction
US9229900B2 (en) 2011-01-20 2016-01-05 Linkedin Corporation Techniques for ascribing social attributes to content
US9674136B2 (en) 2011-01-20 2017-06-06 Linkedin Corporation Techniques for ascribing social attributes to content
US20130227020A1 (en) * 2011-01-20 2013-08-29 Linkedin Corporation Methods and systems for recommending a context based on content interaction
US9247015B2 (en) * 2011-01-20 2016-01-26 Linkedin Corporation Methods and systems for recommending a context based on content interaction
US20120226750A1 (en) * 2011-03-01 2012-09-06 Rex Gibson Online peer review of internet data
US8650149B2 (en) 2011-03-29 2014-02-11 Manyworlds, Inc. Portable inferred interest and expertise profiles
US8645292B2 (en) 2011-03-29 2014-02-04 Manyworlds, Inc. Serendipitous recommendations system and method
US8694457B2 (en) 2011-03-29 2014-04-08 Manyworlds, Inc. Adaptive expertise clustering system and method
US8600926B2 (en) 2011-03-29 2013-12-03 Manyworlds, Inc. Integrated interest and expertise-based discovery system and method
US8676742B2 (en) 2011-03-29 2014-03-18 Manyworlds, Inc. Contextual scope-based discovery systems
US8843433B2 (en) 2011-03-29 2014-09-23 Manyworlds, Inc. Integrated search and adaptive discovery system and method
US8655829B2 (en) 2011-03-29 2014-02-18 Manyworlds, Inc. Activity stream-based recommendations system and method
US8645312B2 (en) 2011-03-29 2014-02-04 Manyworlds, Inc. Expertise discovery methods and systems
US8694442B2 (en) 2011-03-29 2014-04-08 Manyworlds, Inc. Contextually integrated learning layer
US8719213B2 (en) 2011-03-29 2014-05-06 Manyworlds, Inc. Contextually transformed learning layer
US9529915B2 (en) * 2011-06-16 2016-12-27 Microsoft Technology Licensing, Llc Search results based on user and result profiles
US20120323876A1 (en) * 2011-06-16 2012-12-20 Microsoft Corporation Search results based on user and result profiles
WO2013001147A1 (en) * 2011-06-27 2013-01-03 Nokia Corporation Method and apparatus for providing recommendations based on locally generated models
CN102855248A (en) * 2011-06-29 2013-01-02 中国移动通信集团广西有限公司 Determination method, apparatus and system for user characteristic information
CN103842997A (en) * 2011-08-04 2014-06-04 K·波波夫 Searching for and creating an adaptive content
EP2740054A4 (en) * 2011-08-04 2015-10-28 Krasimir Popov Searching for and creating an adaptive content
US8983947B2 (en) * 2011-09-30 2015-03-17 Jive Software, Inc. Augmenting search with association information
US20130086071A1 (en) * 2011-09-30 2013-04-04 Jive Software, Inc. Augmenting search with association information
US8195799B1 (en) * 2011-10-26 2012-06-05 SHTC Holdings LLC Smart test article optimizer
US9594850B2 (en) 2011-11-17 2017-03-14 Sri International Method and system utilizing a personalized user model to develop a search request
US8868590B1 (en) 2011-11-17 2014-10-21 Sri International Method and system utilizing a personalized user model to develop a search request
US20130132851A1 (en) * 2011-11-22 2013-05-23 International Business Machines Corporation Sentiment estimation of web browsing user
CN103136188A (en) * 2011-11-22 2013-06-05 国际商业机器公司 Method and system used for sentiment estimation of web browsing user
US8976955B2 (en) 2011-11-28 2015-03-10 Nice-Systems Ltd. System and method for tracking web interactions with real time analytics
US9900395B2 (en) 2012-01-27 2018-02-20 Comscore, Inc. Dynamic normalization of internet traffic
US8954580B2 (en) 2012-01-27 2015-02-10 Compete, Inc. Hybrid internet traffic measurement using site-centric and panel data
US9734468B2 (en) 2012-02-21 2017-08-15 Nice Ltd. System and method for resolving customer communications
US8977573B2 (en) 2012-03-01 2015-03-10 Nice-Systems Ltd. System and method for identifying customers in social media
US20130238783A1 (en) * 2012-03-09 2013-09-12 Oracle International Corporation Personalized navigation shortcuts
US9311417B2 (en) * 2012-03-09 2016-04-12 Oracle International Corporation Personalized navigation shortcuts
US20150112918A1 (en) * 2012-03-17 2015-04-23 Beijing Yidian Wangju Technology Co., Ltd. Method and system for recommending content to a user
US20130282361A1 (en) * 2012-04-20 2013-10-24 Sap Ag Obtaining data from electronic documents
US9348811B2 (en) * 2012-04-20 2016-05-24 Sap Se Obtaining data from electronic documents
US9628573B1 (en) 2012-05-01 2017-04-18 Amazon Technologies, Inc. Location-based interaction with digital works
US9665630B1 (en) * 2012-06-18 2017-05-30 EMC IP Holding Company LLC Techniques for providing storage hints for use in connection with data movement optimizations
US20140067369A1 (en) * 2012-08-30 2014-03-06 Xerox Corporation Methods and systems for acquiring user related information using natural language processing techniques
US9396179B2 (en) * 2012-08-30 2016-07-19 Xerox Corporation Methods and systems for acquiring user related information using natural language processing techniques
US20140143012A1 (en) * 2012-11-21 2014-05-22 Insightera Ltd. Method and system for predictive marketing campigns based on users online behavior and profile
US9251203B2 (en) * 2012-12-22 2016-02-02 Mmodal Ip Llc User interface for predictive model generation
US9489433B2 (en) * 2012-12-22 2016-11-08 Mmodal Ip Llc User interface for predictive model generation
US20160140193A1 (en) * 2012-12-22 2016-05-19 Mmodal Ip Llc User Interface for Predictive Model Generation
US20140222784A1 (en) * 2012-12-22 2014-08-07 Mmodal Ip Llc User Interface for Predictive Model Generation
US9405746B2 (en) * 2012-12-28 2016-08-02 Yahoo! Inc. User behavior models based on source domain
US20140189525A1 (en) * 2012-12-28 2014-07-03 Yahoo! Inc. User behavior models based on source domain
US9552555B1 (en) * 2012-12-31 2017-01-24 Google Inc. Methods, systems, and media for recommending content items based on topics
US9501506B1 (en) 2013-03-15 2016-11-22 Google Inc. Indexing system
US9483568B1 (en) 2013-06-05 2016-11-01 Google Inc. Indexing system
US20150161292A1 (en) * 2013-12-05 2015-06-11 Richplay Information Co., Ltd. Method for recommending document
US20150193789A1 (en) * 2014-01-03 2015-07-09 Mastercard International Incorporated Method and system for personalized news recommendations based on purchase behavior
US9866999B1 (en) 2014-01-12 2018-01-09 Investment Asset Holdings Llc Location-based messaging
US9430783B1 (en) 2014-06-13 2016-08-30 Snapchat, Inc. Prioritization of messages within gallery
US9825898B2 (en) 2014-06-13 2017-11-21 Snap Inc. Prioritization of messages within a message collection
US9693191B2 (en) 2014-06-13 2017-06-27 Snap Inc. Prioritization of messages within gallery
EP3163465A4 (en) * 2014-06-25 2018-03-14 Beijing Baidupay Science And Tech Co Ltd Method and device for data mining based on user's search behaviour
US9563912B2 (en) 2014-08-15 2017-02-07 Microsoft Technology Licensing, Llc Auto recognition of acquirable entities
WO2016025658A1 (en) * 2014-08-15 2016-02-18 Microsoft Technology Licensing, Llc Auto recognition of acquirable entities
US9444765B2 (en) 2014-11-18 2016-09-13 AO Kaspersky Lab Dynamic categorization of network resources
US9160680B1 (en) 2014-11-18 2015-10-13 Kaspersky Lab Zao System and method for dynamic network resource categorization re-assignment
US9594979B1 (en) * 2014-11-19 2017-03-14 Amazon Technologies, Inc. Probabilistic registration of interactions, actions or activities from multiple views
US9262681B1 (en) * 2014-11-19 2016-02-16 Amazon Technologies, Inc. Probabilistic registration of interactions, actions or activities from multiple views
US9672475B2 (en) 2014-12-19 2017-06-06 International Business Machines Corporation Automated opinion prediction based on indirect information
US9697466B2 (en) * 2014-12-19 2017-07-04 International Business Machines Corporation Automated opinion prediction based on indirect information
US9497498B2 (en) * 2015-01-23 2016-11-15 Robert Hain System and method for live streaming of content
US9769208B2 (en) 2015-05-28 2017-09-19 International Business Machines Corporation Inferring security policies from semantic attributes
WO2017035970A1 (en) * 2015-08-31 2017-03-09 北京百度网讯科技有限公司 Information pushing method and apparatus
US9807037B1 (en) 2016-07-08 2017-10-31 Asapp, Inc. Automatically suggesting completions of text
US9715496B1 (en) * 2016-07-08 2017-07-25 Asapp, Inc. Automatically responding to a request of a user
US9805371B1 (en) 2016-07-08 2017-10-31 Asapp, Inc. Automatically suggesting responses to a received message

Also Published As

Publication number Publication date Type
US20080114756A1 (en) 2008-05-15 application
US8825853B1 (en) 2014-09-02 grant
US7685276B2 (en) 2010-03-23 grant
US20100122178A1 (en) 2010-05-13 application
US20060136589A1 (en) 2006-06-22 application
US8301764B2 (en) 2012-10-30 grant
US8626911B2 (en) 2014-01-07 grant
US7320031B2 (en) 2008-01-15 grant
US9098813B1 (en) 2015-08-04 grant
US20130073390A1 (en) 2013-03-21 application

Similar Documents

Publication Publication Date Title
Cooley et al. Web usage mining: discovery and application of interesting patterns from web data
Tang et al. Social recommendation: a review
Mobasher et al. Semantically enhanced collaborative filtering on the web
US7530020B2 (en) Computer graphic display visualization system and method
Xu et al. Privacy-enhancing personalized web search
Huang et al. A graph model for E‐commerce recommender systems
US6314420B1 (en) Collaborative/adaptive search engine
Widyantoro et al. Learning user interest dynamics with a three‐descriptor representation
Bobadilla et al. Recommender systems survey
Anand et al. Contextual recommendation
US20100306249A1 (en) Social network systems and methods
Zhang et al. An associative classification-based recommendation system for personalization in B2C e-commerce applications
Chen et al. PVA: A self-adaptive personal view agent
US6681247B1 (en) Collaborator discovery method and system
US7685198B2 (en) Systems and methods for collaborative tag suggestions
US6591248B1 (en) Banner advertisement selecting method
US20050256866A1 (en) Search system and methods with integration of user annotations from a trust network
Baraglia et al. Dynamic personalization of web sites without user intervention
Anand et al. Intelligent techniques for web personalization
US20020173971A1 (en) System, method and application of ontology driven inferencing-based personalization systems
US5983214A (en) System and method employing individual user content-based data and user collaborative feedback data to evaluate the content of an information entity in a large information communication network
US7328216B2 (en) System and method for personalized search, information filtering, and for generating recommendations utilizing statistical latent class models
US20080005064A1 (en) Apparatus and method for content annotation and conditional annotation retrieval in a search context
Menczer et al. Adaptive retrieval agents: Internalizing local context and scaling up to the Web
US20070106659A1 (en) Search engine that applies feedback from users to improve search results

Legal Events

Date Code Title Description
AS Assignment

Owner name: UTOPY, INC., CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:KONIG, YOCHAI;BERTHOLD, MICHAEL R.;TWERSKY, ROY;REEL/FRAME:010900/0147

Effective date: 20000615

AS Assignment

Owner name: BLACKSMITH VENTURES I-A, L.P., CALIFORNIA

Free format text: SECURITY AGREEMENT;ASSIGNOR:UTOPY, INC.;REEL/FRAME:015410/0076

Effective date: 20041124

Owner name: BLACKSMITH CAPITAL UTP L.L.C., CALIFORNIA

Free format text: SECURITY AGREEMENT;ASSIGNOR:UTOPY, INC.;REEL/FRAME:015410/0076

Effective date: 20041124

AS Assignment

Owner name: SQUARE 1 BANK, NORTH CAROLINA

Free format text: SECURITY AGREEMENT;ASSIGNOR:UTOPY, INC.;REEL/FRAME:017097/0220

Effective date: 20060131

AS Assignment

Owner name: PERSONALIZED USER MODEL (PUM), NEW YORK

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:LEVINO;REEL/FRAME:020387/0123

Effective date: 20070523

Owner name: LEVINO, CYPRUS

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:UTOPY, INC.;REEL/FRAME:020381/0042

Effective date: 20060303

FPAY Fee payment

Year of fee payment: 4

RR Request for reexamination filed

Effective date: 20110314

FPAY Fee payment

Year of fee payment: 8

FPAY Fee payment

Year of fee payment: 12